Increased Enthesopathy Among Behçet s Syndrome Patients With Acne and Arthritis

Transcription

1 ARTHRITIS & RHEUMATISM Vol. 58, No. 5, May 2008, pp DOI /art , American College of Rheumatology Increased Enthesopathy Among Behçet s Syndrome Patients With Acne and Arthritis An Ultrasonography Study Gulen Hatemi, Izzet Fresko, Koray Tascilar, and Hasan Yazici Objective. Behçet s syndrome is characterized by clusters of disease expression, one of which is the coexistence of acne and arthritis. The aim of this study was to test the hypothesis that enthesopathy is increased in this cluster, having features reminiscent of acneassociated arthritis. Methods. The study group comprised 35 patients with Behçet s syndrome who had acne and arthritis (BS-AA), 38 patients with Behçet s syndrome who did not have arthritis (BS-WA), 37 patients with ankylosing spondylitis (AS), 25 patients with rheumatoid arthritis (RA), and 25 healthy control subjects. Five entheseal sites (the superior and inferior poles of the patella, the tibial tuberosity, and the superior and inferior poles of the calcaneus) on both lower extremities were examined by 2 independent observers, using ultrasonography. A previously validated composite score was calculated for each patient. The numbers of entheseal sites giving a positive power Doppler signal in each group were also compared. Results. Patients with AS had the highest enthesopathy score (mean SD ; F [4df] 8.43, P < 0.001) followed by BS-AA patients ( ; F [3df] 3.63, P 0.015). The mean SD enthesopathy scores among the remaining 3 groups were similar (for BS-WA, ; for RA, ; for healthy controls, [F (2df) 0.65, P 0.53]). Power Doppler scores were highest for the BS-AA group Supported by the research fund of Istanbul University (project YÖP-11/ ). Gulen Hatemi, MD, Izzet Fresko, MD, Koray Tascilar, MD, Hasan Yazici, MD: Istanbul University, Cerrahpasa Medical School, Istanbul, Turkey. Address correspondence and reprint requests to Izzet Fresko, MD, Suadiye Yazarlar Sok, Yildiz Apt 12/6, Istanbul, Turkey izzetfresko@yahoo.com. Submitted for publication July 18, 2007; accepted in revised form February 1, (mean SD ; F [4df] 15.54, P < 0.001) followed by the AS group ( ; F [3df] 14.38, P < 0.001), the RA group ( ; F [2df] 13.88, P < 0.001), the BS-WA group ( ; F [1df] 4.48, P 0.038), and the control group ( ). There was 87% agreement between the 2 observers ( 0.55, intraclass correlation coefficient 0.71). Conclusion. The increased presence of enthesopathy among BS-AA patients compared with that among BS-WA patients further supports the hypothesis that patients with Behçet s syndrome who also have arthritis and acne form a distinct cluster. Behçet s syndrome is a vasculitis with protean manifestations. There is evidence that the clinical picture is not homogeneous, and that there are various clusters of disease expression. A factor analysis showed that acne and arthritis, superficial thrombophlebitis and deep vein thrombosis, oral ulcers along with genital ulcers, and erythema nodosum are examples of such clusters (1). The coexistence of acne and arthritis suggests that a subgroup of patients with Behçet s syndrome may have acne-associated arthritis, which can be evaluated within the broader concept of seronegative spondylarthropathies. Enthesitis is an important clinical feature of the spondylarthropathies. An increased presence of enthesitis in a group of Behçet s syndrome patients with acne and arthritis (BS-AA) would support an association with spondylarthropathies in this subset. Thus, we studied the frequency of enthesopathy in BS-AA patients and compared it with the frequency among Behçet s syndrome patients without arthritis (BS-WA), patients with ankylosing spondylitis (AS), patients with rheumatoid arthritis (RA), and healthy control subjects, using B-mode and power Doppler ultrasonography. 1539

2 1540 HATEMI ET AL Figure 1. Top left, Erosion at the Achilles tendon insertion on the superior pole of the calcaneus. Top right, Erosion at the insertion site of plantar facsia to the inferior pole of the calcaneus. Bottom, Power Doppler signals and an enthesophyte at the superior pole of the calcaneus. PATIENTS AND METHODS Patients. Patients with Behçet s syndrome who fulfilled the International Study Group for Behcet s Disease criteria for a diagnosis of Behçet s disease (2) were surveyed. Patients with AS and patients with RA who were followed up in the rheumatology outpatient clinic of the same unit and healthy subjects were used as controls. To identify Behçet s syndrome patients with acne and arthritis, we searched our Behçet s syndrome patient database for those who had reported having both acne and arthritis. We identified 236 such patients among the 1,488 patients who were registered at our clinic with a diagnosis of Behçet s syndrome between 2000 and We then reviewed the charts of these patients to identify those

3 ENTHESOPATHY IN BEHÇET S SYNDROME PATIENTS WITH ACNE AND ARTHRITIS 1541 Figure 2. Mean (95% confidence interval [95% CI]) enthesopathy scores and power Doppler scores for each group. BSWA Behçet s syndrome without arthritis; BSAA Behçet s syndrome with acne and arthritis; AS ankylosing spondylitis; RA rheumatoid arthritis; HC healthy control. whose acne and arthritis was observed and documented by one of the members of our Behçet s syndrome research group. Patients were contacted by telephone, and those who agreed to participate in the study, mainly based on geographic convenience, were included. Having acne and arthritis at the time of ultrasound examination was not a requirement. Patients with Behçet s syndrome who did not have arthritis, patients with AS, and patients with RA were recruited from among those attending the clinic for their routine visits to the clinic. Subjects younger than age 18 years and those who had previous knee or ankle joint surgery were excluded. The study was approved by the ethics committee of Istanbul University, Cerrahpasa Medical School. Ultrasound examination. B-mode ultrasonography was performed in a masked manner by 2 independent rheumatologists (GH and IF) on the same day. Patients lay supine for examination of the knee and prone with their feet hanging down the side of the bed for examination of the heel. A Shimadzu 1200 device (Shimadzu, Columbia, MD) with a linear probe of 8 15 MHz frequency was used. Five entheseal sites were scanned on both legs, as follows: the superior pole of the patella at the insertion of the quadriceps tendon, the inferior pole of the patella at the proximal insertion of the patellar tendon, the tibial tuberosity at the distal insertion of the patellar tendon, the superior pole of the patella at the insertion of the Achilles tendon, and the inferior pole of the calcaneus at the insertion of the plantar fascia. These sites were examined for erosions, enthesophytes, tendon thickness, and bursitis. An enthesophyte was defined as a step-up prominence on a normal bone contour, and erosion was defined as a cortical defect with a step-down contour (Figure 1). Bursitis was defined as a well-circumscribed anechoic or hypoechoic area with a longitudinal diameter of 2 mm, at the site of an anatomic bursa. Tendon thickness was measured at the point of maximal thickness proximal to the bony insertion. Measurements 6.1 mm for the quadriceps tendon, 4 mm for the patellar tendon, 5.3 mm for the Achilles tendon, and 4.4 mm for the plantar fascia were defined as abnormal (3). The cutoff limits for tendon thickness were also calculated by adding 2 SD to the mean values for the healthy control subjects. Scores were calculated for both sets of values. The Glasgow Ultrasound Enthesitis Scoring System (GUESS) was used to calculate the enthesopathy score for each patient (3). One point was given for each entheseal abnormality (thickened tendon, bursitis, erosion, and enthesopathy) (Figure 2). The maximum possible score was 36. The average total score from the 2 observers was used in the calculations. Power Doppler ultrasound examination. Power Doppler ultrasound examination was performed after B-mode ultrasonography at each entheseal site. The settings were standardized with a pulse repetition frequency of 750 MHz, and the gain was set to a point at which no signal was detected under the bone. One point was given for each entheseal site with a signal, and a cumulative score (maximum possible 10) was obtained. Clinical examination. The same entheseal sites were clinically evaluated by a rheumatologist on the same day after the ultrasound examination, and any pain, tenderness, or swelling was noted. Statistical analysis. The enthesopathy and power Doppler scores for the groups were compared by analysis of variance (ANOVA). Nonparametric values were evaluated by the Kruskal-Wallis or the chi-square test. A t-test for independent samples was used for two-way comparisons. Interobserver reliability was assessed using the kappa test and the intraclass correlation coefficient. RESULTS Clinical characteristics of the study subjects. Thirty-five BS-AA patients, 38 BS-WA patients, 37 patients with AS, 25 patients with RA, and 25 healthy control subjects were studied. The demographic characteristics of the patients are shown in Table 1. There were significantly more women among the group of patients with RA compared with the other groups ( 2 [4df] , P 0.001; when RA is excluded, 2 [3df] 6.881, P 0.07, by Kruskal-Wallis test). Patients with

4 1542 HATEMI ET AL Table 1. Demographic characteristics of the patients and control subjects* Subject group (no.) No. men/ no. women Age, mean SD years BS-WA (38) 24/ BS-AA (35) 25/ AS (37) 32/ RA (25) 4/ Controls (25) 15/ * BS-WA Behçet s syndrome without arthritis; BS-AA Behçet s syndrome with acne and arthritis; AS ankylosing spondylitis; RA rheumatoid arthritis. RA were significantly older than patients in the other groups (F [4df] 6.918, P 0.001, by ANOVA), while the ages of patients in the other groups were similar (F [3df] 0.515, P 0.67). There were no differences in age or sex between the BS-AA and BS-WA groups. The clinical features of Behçet s syndrome patients that were recorded in the patient charts are shown in Table 2. Ultrasound examination revealed arthritis of the knees or ankles in 4 patients with BS-AA, 4 patients with AS, and 2 patients with RA. The presence of acne was not specifically sought at the time of the ultrasound examination. All of the patients with BS-AA, 14 (38%) of the patients with AS, and 20 (80%) of the patients with RA had a history of arthritis in the knee and/or ankle joints. The mean SD disease duration was years for patients with BS-AA and years for patients with BS-WA. BS-AA patients reported having both acne and arthritis for a mean SD duration of years, and these lesions had first been documented by our group years before the patients were recruited to this study. Papulopustular lesions were more common, whereas deep vein thrombosis was less frequent among BS-AA patients compared with BS-WA patients (P 0.03 and P 0.03, respectively). None of the BS-WA patients had a history of arthritis (Table 2). Enthesopathy evaluation. The enthesopathy score for each group according to GUESS is shown in Figure 2. Among the 5 groups studied, patients with AS had the highest enthesopathy score ( ; F [4df] 8.43, P 0.001), followed by patients with BS-AA ( ; F [3df] 3.63, P 0.015). The mean SD enthesopathy scores among the remaining 3 groups were similar (for BS-WA, ; for RA, ; for control subjects, [F (2df) 0.65, P 0.53]). Intergroup comparisons revealed that BS-AA patients, while having lower enthesopathy scores than AS patients (t 1.96, P 0.05), had higher scores than BS-WA patients (t 3.17, P 0.002). When the enthesopathy score was recalculated using the cutoff values for tendon thickness obtained from our healthy control subjects, the findings remained robust. The mean SD score was highest among patients with AS ( ; F [4df] 11.08, P 0.001) followed by patients with BS-AA ( ; F [3df] 3.93, P 0.01). BS-WA patients, patients with RA, and healthy control subjects had similar scores (mean SD , , and , respectively; F [2df] 2.53, P 0.09). BS-AA patients had lower scores than patients with AS (t 2.94, P 0.004) and higher scores than BS-WA patients (t 3.491, P 0.001). The number of patients with soft tissue and bony components of enthesopathy is shown in Table 3. The number of patients with the soft tissue component of enthesopathy was highest among those with BS-AA ( 2 [4df] 19.63, P by Kruskal-Wallis test), and the number of patients with the bony component of enthesopathy was highest among those with AS ( 2 [4df] 15.4, P 0.004). Power Doppler scores. The mean SD power Doppler score was highest in the BS-AA group ( ; F [4df] 15.54, P 0.001), followed by the AS group ( ; F [3df] 14.38, P 0.001), the RA group ( ; F [2df] 13.88, P 0.001), the BS-WA group ( ; F [1df] 4.48, P 0.038), and healthy control subjects ( ) (Figure 2). Intergroup comparisons showed that the power Doppler score for BS-AA patients was significantly higher than that for BS-WA patients (t 5.85, P 0.001) and healthy control subjects (t 7.23, P 0.001). Comparison of clinical and ultrasound findings. A total of 1,600 entheseal sites in 160 patients were examined. The number of patients with painful/tender and swollen entheseal sites is shown in Table 4. The control group had the smallest number of subjects with painful/tender entheseal sites ( 2 [4df] , P 0.001, by Kruskal-Wallis test), followed by the BS-WA group ( 2 [3df] , P 0.004, by Kruskal-Wallis Table 2. Clinical features of Behçet s syndrome patients with acne and arthritis (BS-AA) and those without arthritis (BS-WA)* Characteristic BS-AA (n 35) BS-WA (n 38) P Oral ulcers 35 (100) 38 (100) 1.0 Genital ulcers 30 (86) 28 (74) 0.16 Papulopustular lesions 35 (100) 33 (87) 0.03 Erythema nodosum 20 (57) 16 (42) 0.15 Arthritis 35 (100) Deep vein thrombosis 2 (6) 9 (24) 0.03 Arterial aneurysms 1 (3) 1 (3) 1.0 Eye involvement 16 (46) 18 (47) 1.0 Neurologic involvement 0 0 Pathergy positivity 25 (71) 24 (63) 0.31 * Values are the number (%). All data were recovered from the patients charts.

5 ENTHESOPATHY IN BEHÇET S SYNDROME PATIENTS WITH ACNE AND ARTHRITIS 1543 Table 3. Number (%) of patients and control subjects with soft tissue and bone components of the Glasgow Ultrasound Enthesitis Scoring System* Subject Component of enthesopathy group (no.) Soft tissue Bone BS-WA (38) 12 (32) 7 (18) BS-AA (35) 27 (71) 14 (40) AS (37) 22 (60) 20 (54) RA (25) 10 (40) 7 (28) Controls (25) 9 (36) 4 (16) * BS-WA Behçet s syndrome without arthritis; BS-AA Behçet s syndrome with acne and arthritis; AS ankylosing spondylitis; RA rheumatoid arthritis. P versus all other groups, by Kruskal-Wallis test. P versus all other groups, by Kruskal-Wallis test. test). The number of patients with painful/tender entheseal sites was similar in the remaining 3 groups ( 2 [2df] 2.831, P 0.243, by Kruskal-Wallis test). None of the healthy control subjects or BS-WA patients had swollen entheseal sites. The other 3 groups had similar numbers of patients with swollen entheseal sites ( 2 [2df] 0.017, P 0.992, by Kruskal-Wallis test). There was a significant correlation between ultrasound findings and both pain/tenderness (r 0.281, P 0.001) and swelling (r 0.109, P 0.001) on physical examination. Interobserver reliability. For the 7,360 items evaluated, agreement between the 2 observers was 87% ( 0.55, intraclass correlation coefficient 0.71). DISCUSSION Our results showed that patients with BS-AA had more enthesopathy compared with BS-WA patients. Seventy-one percent of BS-AA patients had a soft tissue enthesitis (tendon thickening and/or bursitis), and 40% had a bony component of enthesitis (erosion and/or enthesophyte); these values for BS-WA patients were 32% and 18%, respectively (Table 3). Acne-associated arthritis is a clinical entity that was first described by Windom et al in 1961 (4). Since then, several case series with this association have been reported (5 8). Acne fulminans, acne conglobata, and hidradenitis suppurativa are the skin lesions most commonly associated with joint disease. Clinical features include chronic episodic peripheral arthritis that is usually oligoarticular and affects large joints, axial involvement, and enthesopathy (9). Radiologic features were described as indistinguishable from those of reactive arthritis and psoriatic arthropathy (7). The lack of an association with HLA B27 was reported (9). The pathogenesis of acne-associated arthritis is not known. However, a reactive arthritis like mechanism including an inappropriate response to bacterial antigens involved in acne, as well as genetic and autoimmune mechanisms have been postulated (8). Previously, Diri et al examined the relationship between acne and arthritis in Behçet s syndrome (10). In that study, 44 patients with Behçet s syndrome and active arthritis, 42 patients with Behçet s syndrome and no history of arthritis, 21 patients with RA, and 33 healthy volunteers were examined for the presence of comedones, papules, and pustules. The number of papules and pustules was higher among the Behçet s syndrome patients with arthritis compared with control subjects. In another study, which explored target organ associations in Behçet s syndrome among a different group of patients, factor analysis identified 4 clusters of disease expression, among which was also the association of acne and arthritis, again confirming our previous observations (1). Our group also showed that pustular acneiform lesions in Behçet s syndrome were not sterile Table 4. Clinical and ultrasonographic entheseal findings in each group* BS-WA (n 38) BS-AA (n 35) AS (n 37) RA (n 25) Controls (n 25) P Clinical examination Tender/painful entheses, no. (%) 10 (26) 24 (69) 20 (54) 12 (48) 3 (12) Swollen entheses, no. (%) 0 6 (17) 6 (16) 4 (16) Components of enthesopathy score (GUESS) per subject Tendon thickening Bursitis Erosions Enthesophytes * Except where indicated otherwise, values are the mean SD. BS-WA Behçet s syndrome without arthritis; BS-AA Behçet s syndrome with acne and arthritis; AS ankylosing spondylitis; RA rheumatoid arthritis; GUESS Glasgow Ultrasound Enthesitis Scoring System. P values are comparing the number of subjects with tender/painful and swollen entheses in each group and the GUESS score for each component in each group.

6 1544 HATEMI ET AL (11). Furthermore, prophylactic penicillin treatment was shown to be effective for both the mucocutaneous lesions and the arthritis episodes in Behçet s syndrome (12). Current data, along with these observations, suggest that arthritis in Behçet s syndrome may be related to acne-associated arthritis. Our findings in a group of patients with Behçet s syndrome bring us back to the discussion regarding whether Behçet s syndrome belongs to the spondylarthropathies. The issue of sacroiliitis associated with Behçet s syndrome had been widely explored. Some investigators suggested that the frequency of sacroiliitis (and actually AS) is increased among patients with Behçet s syndrome and thought that it belonged to the spondylarthropathy complex (13), while other investigators, including our group, observed that this was not the case (14). However, to our knowledge, neither the studies supporting the coexistence of spondylarthropathy and Behçet s syndrome nor those presenting evidence otherwise specifically involved a subgroup of patients with Behçet s syndrome who also had acne and arthritis. Another issue is enthesopathy without sacroiliitis in Behçet s syndrome. The only study specifically designed to search for enthesopathies used physical examination and radiography, and that study showed that 5 of 174 patients with Behçet s syndrome had enthesopathies (15). The highest frequency of enthesopathy was reported by Caporn et al, in their study in which they examined the hospital records of 14 patients with Behçet s syndrome (16). Thirteen of those patients had joint involvement, and 5 (36%) had enthesopathy. All 5 patients with enthesopathy had calcaneal spurs, and 1 of them also had vertebral body squaring. Chang et al (17) reported 2 patients with enthesitis among their 58 patients with Behçet s syndrome (3.4%). In their study, enthesopathy was evaluated by physical examination. Kim et al (18) evaluated 59 patients with Behçet s syndrome who had joint involvement. Radiologic examination was performed in 42% of these patients, and enthesopathy (bony spurs) was detected in 3 of them (18). In a retrospective observation, Imbert et al reported 1 patient with tendinitis among 65 patients with Behçet s syndrome, 53% of whom had arthritis (19). Among the 47 Behçet s syndrome patients with arthritis reported by Yurdakul et al, 2 had Achilles tendinitis, and 1 had calcaneal erosions (20). This wide variation in the reported frequency of enthesopathy in Behçet s syndrome may be attributable to the fact that radiography and physical examination are not sensitive methods for detecting enthesopathy and also to the different populations studied, with different frequencies of joint involvement. Moreover, none of these studies was performed in a subgroup of BS-AA patients. This study is the first to extensively explore enthesopathy in patients with Behçet s syndrome, using ultrasonography. The only other study utilizing musculoskeletal ultrasound in Behçet s syndrome was performed by Ozcakar et al (21), and that study showed that the quadriceps tendon thickness in patients with Behçet s syndrome is increased compared with that in patients with familial Mediterranean fever and healthy control subjects, and that Achilles tendon thickness is increased compared with that in healthy control subjects. The authors stated that tendon thickness is not related to the presence of arthritis. The number of patients with arthritis studied in each group, however, was small. Previous studies have shown that ultrasonography has a higher sensitivity than clinical examination for detecting enthesopathy. D Agostino and colleagues reported that careful clinical examination of entheseal sites revealed abnormalities in 14.4% of the sites in 34 patients with spondylarthropathy, while results of ultrasound examination were abnormal in 36% of the entheseal sites (22). Balint et al reported the frequency of tender entheseal sites as 20% and that of swollen entheseal sites as 4% in 35 patients with spondylarthropathy, whereas abnormal entheseal findings were detected by ultrasonography in 56% of these patients (3). GUESS is the only available validated ultrasound-based scoring system for enthesopathy. It is an efficient tool, but it evaluates tendon and ligament pathology based solely on the presence or absence of thickness while leaving out hypoechoic changes. As Balint et al (3) also stated, this may cause underestimation of enthesitis. Even if this were the case, it would be true for all groups and thus most probably would not appreciably affect our conclusions. The presence of active arthritis during ultrasound examination may cause false-positive results when evaluating enthesopathy. However, few of the patients in our study had active arthritis during ultrasound examination, and these patients were quite evenly distributed among the BS-AA, AS, and RA groups (4 of 35, 4 of 37, and 2 of 25 patients, respectively). Thus, the presence of active arthritis is unlikely to affect our results. Adding power Doppler ultrasound evaluation adds to the GUESS scoring system by differentiating mechanical enthesopathy and burnt-out entheseal lesions from inflammatory lesions. Power Doppler signals were more frequent among BS-AA patients than among patients with AS. This might be attributable to selection bias. Because only Behçet s syndrome patients whose

7 ENTHESOPATHY IN BEHÇET S SYNDROME PATIENTS WITH ACNE AND ARTHRITIS 1545 arthritis was documented by one of the doctors in our group were included, these patients may have more active disease than AS patients, with perhaps more burnt-out disease. The other problems with power Doppler are the lack of validated and dependable quantification of the signals, the possibility of nutrient arteries causing false-positive signals, and its being highly operator and machine dependent. These problems might have affected our results, but they are unlikely to change our conclusion, because all patients and control subjects were examined by the same device and by 2 independent observers in a masked manner. One other limitation of this study was that ultrasonography could not be tested against a method that was considered the gold standard for detecting enthesopathy. Magnetic resonance imaging is not considered an optimal modality, because changes in the fibrous part of the entheses cannot be easily detected with this method. Moreover, Kamel et al have shown that magnetic resonance imaging is less sensitive than ultrasonography in detecting both knee and heel enthesitis, because it cannot detect calcifications at insertion sites (23,24). Our results confirmed our contention that the frequency of enthesopathy is increased among patients with Behçet s syndrome who also have acne and arthritis compared with Behçet s syndrome patients without this association. The increased presence of enthesopathy among BS-AA patients compared with BS-WA patients further supports the hypothesis that BS-AA patients form a distinct cluster, carrying features of a spondylarthropathy. The presence of HLA B27 and sacroiliitis among this specific subset of patients remains to be studied. AUTHOR CONTRIBUTIONS Dr. Fresko had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study design. Hatemi, Fresko, Yazici. Acquisition of data. Hatemi, Fresko, Tascilar. Analysis and interpretation of data. Hatemi, Fresko, Yazici. Manuscript preparation. Hatemi, Fresko, Yazici. Statistical analysis. Hatemi, Tascilar. REFERENCES 1. Tunc R, Keyman E, Melikoglu M, Fresko I, Yazici H. Target organ associations in Turkish patients with Behcet s disease: a cross sectional study by exploratory factor analysis. J Rheumatol 2002;29: International Study Group for Behcet s Disease. Criteria for diagnosis of Behcet s disease [review]. Lancet 1990;335: Balint PV, Kane D, Wilson H, McInnes IB, Sturrock RD. Ultrasonography of entheseal insertions in the lower limb in spondyloarthropathy. Ann Rheum Dis 2002;61: Windom RE, Sanford JP, Ziff M. Acne conglobata and arthritis. Arthritis Rheum 1961:4: Rosner IA, Burg CG, Wisnieski JJ, Schacter BZ, Richter DE. The clinical spectrum of the arthropathy associated with hidradenitis suppurativa and acne conglobata. J Rheumatol 1993;20: Rosner IA, Richter DE, Huettner TL, Kuffner GH, Wisnieski JJ, Burg CG. Spondyloarthropathy associated with hidradenitis suppurativa and acne conglobata. Ann Intern Med 1982;97: Ellis BI, Shier CK, Leisen JJ, Kastan DJ, McGoey JW. Acneassociated spondylarthropathy: radiographic features. Radiology 1987;162: Thein M, Hogarth MB, Acland K. Seronegative arthritis associated with the follicular occlusion triad. Clin Exp Dermatol 2004; 29: Levin J, Werth VP. Skin disorders with arthritis [review]. Best Pract Res Clin Rheumatol 2006;20: Diri E, Mat C, Hamuryudan V, Yurdakul S, Hizli N, Yazici H. Papulopustular skin lesions are seen more frequently in patients with Behcet s syndrome who have arthritis: a controlled and masked study. Ann Rheum Dis 2001;60: Hatemi G, Bahar H, Uysal S, Mat C, Gogus F, Masatlioglu S, et al. The pustular skin lesions in Behcet s syndrome are not sterile. Ann Rheum Dis 2004;63: Calguneri M, Kiraz S, Ertenli I, Benekli M, Karaarslan Y, Celik I. The effect of prophylactic penicillin treatment on the course of arthritis episodes in patients with Behcet s disease: a randomized clinical trial. Arthritis Rheum 1996;39: Dilsen N, Konice M, Aral O. Why Behçet s disease should be accepted as a seronegative arthritis. In: Lehner T, Barnes CG, editors. Recent advances in Behcet s disease. London: Royal Society of Medicine Services; p Yazici H, Tuzlaci M, Yurdakul S. A controlled survey of sacroiliitis in Behcet s disease. Ann Rheum Dis 1981;40: Hamza M. Enthesitis in Behcet s disease. In: Godeau P, Wechsler B, editors. Behçet s disease. Amsterdam: Elsevier Science Publishers; p Caporn N, Higgs ER, Dieppe PA, Watt I. Arthritis in Behcet s syndrome. Br J Radiol 1983;56: Chang HK, Lee DH, Jung SM, Choi SJ, Kim JU, Choi YJ, et al. The comparison between Behcet s disease and spondyloarthritides: does Behcet s disease belong to the spondyloarthropathy complex? J Korean Med Sci 2002;17: Kim HA, Choi KW, Song YW. Arthropathy in Behcet s disease. Scand J Rheumatol 1997;26: Imbert I, Legros P, Prigent D, Bergaoui N, Sekkat A, Chaouni- Berbich A, et al. Articular manifestations of Behcet s disease: apropos of 65 cases. Rev Rhum Mal Osteoartic 1987;54: Yurdakul S, Yazici H, Tuzun Y, Pazarli H, Yalcin B, Altac M, et al. The arthritis of Behcet s disease: a prospective study. Ann Rheum Dis 1983;42: Ozcakar L, Onat AM, Ureten K, Cetin A, Kiraz S, Ertenli I, et al. Sonographic evaluation of the tendons in familial Mediterranean fever and Behcet s disease. Joint Bone Spine 2006;73: D Agostino MA, Said-Nahal R, Hacquard-Bouder C, Brasseur JL, Dougados M, Breban M. Assessment of peripheral enthesitis in the spondylarthropathies by ultrasonography combined with power Doppler: a cross-sectional study. Arthritis Rheum 2003;48: Kamel M, Eid H, Mansour R. Ultrasound detection of knee patellar enthesitis: a comparison with magnetic resonance imaging. Ann Rheum Dis 2004;63: Kamel M, Eid H, Mansour R. Ultrasound detection of heel enthesitis: a comparison with magnetic resonance imaging. J Rheumatol 2003;30:774 8.