Osteonecrosis as a rare musculoskeletal complication in Behcet’s disease- the largest case series with literature review

Abstract

Background

Behcet disease (BD) as a variable vessel vasculitis is mainly characterized by ocular involvement, genital and oral aphthosis, and erythema nodosum. However, major organ involvements including gastrointestinal involvement, nervous system, and vascular involvement are among the severe complications. Osteonecrosis is a rare complication of patients with BD. We aim to report the largest series of BD patients suffering from osteonecrosis.

Methods

We have retrospectively reviewed all patients in Iran Behcet’s Disease Registry and reported those with osteonecrosis. Patients’ medication and clinical features, symptoms, and details of osteonecrosis will also be presented. Furthermore, previously reported cases will also be reviewed.

Results

Seven thousand eight hundred thirty-one patients were diagnosed with BD and registered. 18 patients developed ON with an incidence of 0.22%. The most common involvement during the disease progression was oral aphthosis which appeared in 100% of patients followed by ocular involvement in 85.7% and skin involvement in 71.4%. Vascular, ocular, and nervous system involvements are significantly higher in BD patients with osteonecrosis than the other BD patients. For the management of acute episode of uveitis, deep vein thrombosis, severe gastrointestinal involvement, arterial involvement, nervous system Involvement, and joint involvement high dose of glucocorticoids is indicated.

Conclusions

ON tends to appear as a multifocal involvement in BD patients, hence, after diagnosis of ON in one joint other possible sites of ON should be investigated.

Background

The initial symptoms of Behcet disease (BD) were first noticed by Hippocrates in ancient times [1]. In 1937, it was described as a separate disease by Huluci Behcet [2]. BD, first known as a viral syndrome, today is categorized as variable vessel vasculitis [1, 2]. Oral and genital aphthosis, erythema nodosum, arthritis, and ocular involvement are among the most common manifestation [1]. BD often follows a relapsing-remitting pattern; however, some of the organ involvement including the eye, vascular, gastrointestinal, and nervous system could result in severe and irreversible complications [3]. Osteonecrosis is a disturbing complication that has been frequently reported in lupus patients and is strongly associated with corticosteroid therapy. The pathologic pathway of corticosteroid-induced osteonecrosis is through suppression of both osteoblast and osteoclast and increasing apoptosis of osteoblast. Other suggested pathologic mechanisms of osteonecrosis include fat embolism, altered lipid metabolism, impaired repair of microfractures, primary cell death, vasculitis, and vasospasm [4]. Corticosteroid therapy is indicated in the major organ involvement and the flare of the disease [5]. The number of BD patients with osteonecrosis is uncommon in comparison with other rheumatologic diseases. Here we report the clinical features of 14 patients with BD who experienced osteonecrosis (ON). A literature review was also performed and previously published case reports are also discussed.

Patients and methods

From 1975 to 2022 all patients diagnosed with BD were gathered and registered in Iran Behcet’s Disease Registry. The patients were diagnosed by expert opinion, even though most of them fulfilled one of the main classification criteria for BD. The patients’ demographic data in the data registry has already been discussed [6]. The diagnosis of ON was based on imaging (MRI and plain radiographs) and clinical symptoms and classified based on Ficat-Arlet staging. Patients’ medical records and charts were reviewed to look for demographic data, duration of follow-up and diagnosis, first manifestations, organ involvement, and lab test. Details of corticosteroid therapy and treatment regimen were also investigated.

Search strategy

We have searched the Medline database from 9 May 1981 until 9 November 2022 to find manuscripts investigating osteonecrosis in patients with BD. The search strategy was built from combination of ‘Behcet’ and ‘osteonecrosis’ and related Mesh terms (Supplementary file 1). Case reports, randomized clinical trials, cohort, cross-sectional, case-control, and letters were included. Exclusion criteria were non-English manuscripts and missing data. After screening, one case series and six case reports were included [7,8,9,10,11,12,13]. Furthermore, reference lists of included manuscripts were searched by hand to ensure all relevant studies are included. In the final analysis 23 patients from 12 studies were included [7,8,9,10,11,12,13,14,15,16,17,18]. Summary of included studies are shown in Tables 1 and 2.

Table 1 Summary of previously reported and the present study BD patients with ON

Table 2 Summary of corticosteroid therapy BD patients with ON

Statistical analysis

Statistical analysis was conducted using SPSS version 23 for windows (IBM, Armonk, New York). Continuous variables are reported as mean ± sd. Categorical variables are reported as numbers and percentages.

Results

Demographic data and clinical manifestations

Seven thousand eight hundred thirty-one patients were diagnosed with BD and registered. The incidence of ON in our dataset is 0.22%. All of them fulfilled the International Criteria for Behcet’s Disease (ICBD) classification criteria for BD. Three patients were lost to follow up. One patient with multifocal osteonecrosis has already been reported [16]. 14 patients remained for further report including 13 males and one female. The mean age at the onset of the disease was 25.14 ± 6.64 years (range 15–36). The patients’ demographic data are shown in Table 1.

The first manifestation was oral aphthosis in all patients except one who presented with uveitis. The most common involvement during the disease progression was oral aphthosis which appeared in 100% of patients followed by ocular involvement in 85.71% and skin involvement in 71.43%. In those with skin and ocular involvement, pseudofolliculitis and pan ophthalmitis were prevalent, respectively.

HLA-B51 was presented in 35.71% of patients. In none of the cases, the antiphospholipid antibody was detected and the result of the pathergy skin test was positive in 50% of patients.

Corticosteroid therapy

The mean cumulative dosage of corticosteroid therapy received by patients before ON is 13,377.14 ± 11,462.30 mg (range 3,560–45,430). The highest daily dosage of corticosteroid therapy ranged from 30 to 75 mg. Table 2 demonstrates details of corticosteroid therapy. Osteoporosis, cushingoid condition, and hyperlipidemia were investigated as the complications of corticosteroid therapy. Osteoporosis as the most common complication occurred in 87.5% of patients.

ON

Twenty-five joints were involved in 14 patients. In 13 patients, the hip was the only joint complicated with ON and in one patient ON occurred in both shoulders and both hips. Nine patients developed bilateral hip osteonecrosis. Eleven patients were diagnosed with ON at stage III of the disease based on Ficat-Arlet staging. The mean duration between the first symptoms of BD and pain in the involved joint was 89.29 ± 36.96 (range 41–162) months. However, BD and ON are diagnosed with a delay of 74.21 ± 35.97 months and 6.71 ± 5.57 months respectively. This results in a span of 34.36 ± 37.87 months between the diagnosis of BD and the diagnosis of ON. The mean Duration of steroid therapy before development of osteonecrosis was 28 ± 31.6 months. No symptoms of arthritis were detected before the ON in the involved joints (Table 3).

Table 3 Patients’ data pertained to osteonecrosis

Medication

Among immunosuppressant drugs in the management of BD azathioprine, methotrexate, and cyclophosphamide pulse were used in 57.14% of patients each. Alendronate was used in 64.28% of patients (Table 1).

Discussion

In comparison with the previously published case series, we have reported the highest number of BD patients with ON. The mean age of all patients (including our dataset and the literature) is 33.08 years. Of the 36 patients (one study didn’t provide information regarding patient’s sex) eight patients are female.

As the cumulative dose of exogenous corticosteroids increases, the probability of osteonecrosis increases dramatically. This could be due to prolonged exposure or pulses [19]. The cumulative dosage of corticosteroids exceeds 15 gr in six of 14 patients in our dataset. Three of them were because of pulse therapy and two were exposed to corticosteroids for more than 7 years. The last one had the second-highest dosage of corticosteroid usage per day. Cumulative dosage of corticosteroids was more than 15gr in five of the ten patients in the literature. However, in the study by Chang et al. a patient developed ON and no history of corticosteroid therapy was reported. In our dataset three patients required corticosteroid pulses while in the literature 10 of the 13 patients had corticosteroid pulses before development of ON. Patients develop ON on an average of 30.62 months after first exposure to corticosteroids. The odds ratio of developing osteonecrosis in patients with high cumulative dosage of corticosteroid therapy (more than 10g) is 2.4 (95% CI. 0.8 to 6.4) and osteonecrosis occur in 6.7% of patients with high dosage of corticosteroid therapy [20]. Although 50 percent of BD patients experience ocular involvement, and corticosteroid therapy is a common regimen specifically in the flare of the disease the incidence ON in BD patients remains low [5].

In our case series proportion of ocular (85.7%), vascular (35.7%), and neurologic (28.6%) involvements are significantly higher than other patients with BD [1]. In the Iran data registry, the incidence of ocular, vascular, and neurologic involvements are 56.8%, 8.3%, and 3.7% respectively [1]. Osteonecrosis could be due to the high dosage of corticosteroids which is required to overcome these complications [5]. In the literature the vascular (30.43%), Ocular (34.78%), and NS (26.1%) were the main organ involvements. Vascular involvement in BD may also play a role in progression of ON, however, there is not enough evidence to support this hypothesis [21]. In the study by Elgengehy et al. Vasculitis damage index in BD patients is significantly associated with ON [22]. Vascular damage and impaired endothelial dysfunction play different role in development of ON [23]. So cumulative effect of vasculitis and high dose corticosteroid in these patients may be considerable in ON occurrence. In the literature 26.1% of patients experienced NS involvement. In our dataset 28.57% of patients suffered from NS involvement, while, the proportion of patients with NS involvement in Iran data registry is approximately 3% and in Germany Registry, Turkey Cohort, ICBD 27 countries, and Inssbruck Cohort is lower than 24% [1]. In a cohort study in 2021 nervous system (NS) involvement was the independent risk factors of osteonecrosis in lupus patients [24].

The most common site of osteonecrosis was hip involvement which occurred in 27 of the 37 patients. In the literature 26.08% of patients had bilateral hip involvement and 43.48% of patients suffered from bilateral knee ON. The incidence of unilateral hip involvement in the literature is 39.13%. In our dataset, all patients had ON of the hip joint and in 64.28% of cases, the ON appeared as a bilateral involvement [7]. In patients with secondary induced osteonecrosis, 64% of cases with atraumatic osteonecrosis of the knee and 55% of those with atraumatic osteonecrosis of the hip joint are bilateral cases [25, 26]. Based on Ficat-Arlet classification 71.43% of our patients diagnosed with stage III of osteonecrosis of the hip joint which is comparable with cases suffering from secondary induced osteonecrosis of the hip joint [25]. Mont et al. reported a weighted mean time of 49 months (range two-143 months) from diagnosis of asymptomatic disease to collapse of the femoral head [27]. Furthermore more than 50% of early-stage osteonecrosis of the hip and 96% of the stage III of osteonecrosis of the hip will progress to collapse of the hip joint [27].

Limitation and strengths

Since the number of BD patients with ON is very low in comparison with whole data registry, we were not able to conduct an analysis due to random error. Also, three patients were lost to follow-up which is a considerable number in comparison with our data set (14 patients). However, our study is the largest reported case series of BD patients with ON and almost every clinical feature of patients and ON are reported and discussed. The mechanism of ON in BD patients is not completely understood and further longitudinal studies are needed to prove the role of corticosteroids and vascular complications or other risk factors in the development of ON in BD patients.

Conclusion

Corticosteroid therapy is the main risk factor of ON in BD patients. ON tend to appear as a multifocal involvement in BD patient, hence, after diagnosis of ON in one joint other possible sites of ON should be investigated. Vascular involvement is one of the most prevalent major organ complications in BD patients. The mechanism of vascular involvement in osteonecrosis of BD patients and the cumulative effect of corticosteroid therapy should be clarified in the future.