Background

Human T-lymphotropic virus type 1 (HTLV-1) was the first identified human retrovirus [1]. Four decades later, the natural course of HTLV-1 infection and the clinic manifestations of this viral infection are not fully known. An estimated 10–20 million individuals are currently infected with HTLV-1, with higher concentrations found in endemic regions of developing countries, such as Brazil [2, 3]. The two main diseases associated with HTLV-1 infection are adult T-cell leukemia/lymphoma (ATLL) and HTLV-1 associated myelopathy/tropical spastic paraparesis (HAM/TSP), which occur in less than 5% of infected individuals [4, 5]. Nonetheless, evidence has been accumulated that HTLV-1 infection is associated with other diseases and may present a variety of clinical manifestations, including overactive bladder, dry syndrome, infective dermatitis, polymyositis, periodontal disease, arthropathy and erectile dysfunction [5,6,7].

HTLV-1-associated arthropathy was initially attributed to the occurrence of autoimmunity in infected individuals [8, 9], but other studies did not confirm the existence of association between this viral infection with rheumatoid arthritis, Sjögren syndrome or other systemic rheumatic disease [10, 11]. In a few HTLV-1 infected patients with arthritis, atypical lymphocytes were found in the synovial fluid [12], proviral DNA was identified in synovial cells [13] and the presence of Tax protein messenger RNA was documented in the synovial stromal cells [14]. Moreover, transgenic mice expressing the Tax protein develop inflammatory arthropathy [15]. These observations influenced the concept that joint disease in HTLV-1 was associated with a chronic synovitis as observed in rheumatoid arthritis [12]. However, from a clinical perspective, polyarthralgia, not arthritis, is the more relevant finding in HTLV-1-infected patients with joint manifestations [7, 16]. Actually, edema/warm was only observed in 23% of HTLV-1 infected subjects who had chronic joint pain [16]. Moreover, to date, no diagnostic criteria for HTLV-1 associated arthropathy (HAAP) has been established and the radiographic features of the joints in such patients have not been described. The aim of this study was to describe clinic and radiographic features of HTLV-1 infected subjects with joint pain, to better define HTLV-1 associated arthropathy.

Methods

This study was performed in the HTLV-1 outpatient clinic of the Federal University of Bahia Hospital, Salvador, Brazil, from 2016 to 2018. This is a cross sectional study comparing clinic and radiographic features of HTLV-1 infected patients of both gender with age between 18 and 75 years who had joint pain for at least 01 year. Of the 104 HTLV-1 infected subjects who complained of joint pain, 23 were excluded, 15 due to incomplete radiographic data, and 8 who had diagnosis of systemic rheumatic disease under the American College of Rheumatology (ACR) criteria [17]. Additional exclusion criteria were previous history of joint surgery and contraindication to perform conventional radiographic. Participants were 81 HTLV-1-infected individuals and a control group consisted of 30 HTLV-1 seronegative individuals with hip, knee or ankle osteoarthritis followed in the Rheumatology clinic and evaluated by EMC in a period of 2 months. The diagnosis of osteoarthritis was performed according to the ACR criteria [17]. HTLV-1 infection was diagnosed by ELISA (Cambridge Biotech Corp, Worcester, MA) and confirmed by Western blot (HTLV blot 2.4; Genelab, Singapore). The cases were classified as HTLV-1 carriers and those who had probable or definite myelopathy [18]. HTLV-1 carriers were HTLV-1 infected without evidence of neurologic diseases, probable HAM/TSP were patients who didn’t have motor disability but have sensory signals as urinary dysfunction mainly characterized by overactive bladder [18]. Definite HAM/TSP had spastic paraparesis and Osame motor disability score greater than 1 [18]. All included individuals answered a questionnaire, were independently evaluated by two rheumatologists and had conventional radiographs of the hips in anteroposterior (AP) incidence and of the knees and ankles in AP and lateral incidences. These images were independently analyzed by two blind radiologists regarding the diagnosis of the arthropathy. In the case of discrepancies between radiological evaluations, a third radiologist was consulted. The joints of the hips, knees and ankles were selected to be evaluated due to great frequency of involvement in accordance with complaints of HTLV-1 infected individuals, and due to the common involvement of these joints in patients with osteoarthritis [19]. The assessment of joints using conventional radiography was performed due to low cost, abundant availability and the possibility of detecting a range of joint alterations. Additionally, magnetic resonance imaging (MRI) of both ankles was performed in 5 HTLV-1 infected patients and in 5 with osteoarthritis who presented enthesophytes on conventional radiography, in a device with a magnetic field of 1.5 T, weighted FSE sequences were performed in T1 and T2, in multiple planes.

This study was approved by the Institutional Review Board of The Federal University of Bahia, resolution number 7/2016, and all participants provided a written informed consent.

The Fisher’s exact and chi-square tests were used to compare among categorical variables, while the Student’s t test was used to compare age between groups. Statistical analysis was performed using SPSS® software version 21 (IBM, USA). Results with p < 0.05 were considered statistically significant.

Results

The demographic characteristics, clinical forms of HTLV-1 infection and patterns of joint pain complaints in the cases and controls are shown on Table 1. No differences were seen between the groups with respect to age and gender. The prevailing pattern of joint pain in the HTLV-1-infected subjects was polyarticular and symmetrical (54%), while the control group predominantly exhibited oligoarticular and asymmetrical pattern (44%) (p < 0.0001). The knee was the joint with more frequent reports of pain in both groups, followed by the ankles and hips.

Table 1 Demographic and clinical characteristics of 81 HTLV-1-infected patients and 30 controls with osteoarthritis
Full size table

The frequency of radiographic changes in the cases and controls is shown on Tables 2 and 3. While the presence of osteophytes and joint space narrowing were the main radiological findings in the control group with osteoarthritis, the frequency of enthesophytes was higher (p˂0.001) in HTLV-1-infected patients than in controls. While joint space reduction and osteophytes were more prevalent in patients with osteoarthritis, the presence of enthesophytes in the absence of joint space reduction or osteophytes was only observed in HTLV-1 infected patients (p 0.001).

Table 2 Frequencies of main radiographic findings in 81 HTLV-1-infected patients and 30 controls with osteoarthritis
Full size table
Table 3 Radiographic findings by joint in 81 patients infected with HTLV-1 and 30 controls with osteoarthritis
Full size table

Further analysis of the three main radiographic findings in each joint indicated higher frequencies of osteophytes in the knees, ankles and hips, as well as joint space narrowing in the knees and ankles in the group with osteoarthritis compared to HTLV-1-infected patients (p 0.02). While no difference was found between the groups regarding the presence of enthesophytes in the knees and hips, this finding was more frequent in the ankles (p = 0.01) of HTLV-1-infected patients than in controls (Table 3). Among the cases, 48 (59%) were carriers, 22 (27.1%) had probable HAM/TSP and 11 (13,5%) had definite HAM/TSP. The polyarticular pattern of joint pain was observed in carriers, probable and definite HAM/TSP. The knee was the main joint involved in all the 3 subgroups of HTLV-1 infected subjects (85%; 88% and 94% respectively; p > 0.05), and frequency of osteophytes in at least one joint was 60%, 54% and 48% in carriers, probable and definite HAM/TSP respectively (p ˃ 0.05). The frequency of enthesophytes without joint space narrowing and without osteophytes was higher in patients with definite HAM/TSP (46%) than in carriers (33%) and in probable HAM/TSP (24%), but these differences did not achieved significance. No difference was observed regarding the radiographic findings between the HTLV-1 carriers and patients with probable or definite HAM. Other less frequent findings included subchondral sclerosis, identified in four HTLV-1-infected individuals and two controls and diffuse osteopenia in three HTLV-1 carriers. Representative radiographic findings in the hip, knee and ankle of one HTLV-1-infected patient are shown in Fig. 1. Enthesophytes in the absence of joint space narrowing and osteophytes are evident in all three joints.

Fig. 1
figure 1

Presence of Enthesopathy Without Joint Space Narrowing or Osteophytes in a 48-year-old Female With HAAP. Enthesophytes (*) seen in iliac crests (A); patellar insertion of quadriceps tendon and patellar and tibial insertions of the patellar tendon (B); calcaneal insertions of calcaneus tendon and plantar fascia (C)

Full size image

MRI of ankles was performed in 5 patients of each group who presented enthesophytes on conventional radiography (Fig. 2). The comparative frequencies of findings in the HTLV-1-positive and osteoarthritis groups were: enthesitis 90% versus 40% (p 0.01), bursitis 70% versus 10% (p 0.006) and osteitis 50% versus 0% (p 0.01), respectively. No difference was observed regarding the frequency of tendinitis between the groups.

Fig. 2
figure 2

MRI of the Ankle of a Patient With HAAP and in a Patient with Osteoarthritis. Liquid distension of the deep retrocalcaneal bursa in a HTLV-1 + patient (A), and in a patient with osteoarthritis (B). Deep retrocalcaneal bursa (*)

Full size image

Discussion

The arthropathies are a group of musculoskeletal diseases affecting the joints, including osteoarthritis and inflammatory disorders as rheumatoid arthritis, psoriatic arthritis, ankylosing spondylitis and reactive arthritis, among others [19,20,21,22]. While neurological impairment associated with HTLV-1 infection, characterized as HAM/TSP, is well-known [4, 5], joint pain despite very frequent is not well studied in HTLV-1-infected patients. While polyarthralgia was observed in 56% of HTLV-1 infected subjects only 27% of age- and gender-matched seronegative individuals complained of these manifestations [16]. Despite cases of HTLV-1-infected patients with severe joint involvement characterized by synovitis have been reported [23], the predominant joint complaint in infected individuals is polyarthralgia, with scant evidence of arthritis [16]. Here HTLV-1 associated arthropathy was clinically characterized by symmetrical polyarthralgia and the main radiographic finding was the occurrence of enthesophytes in the absence of osteophytes and joint space narrowing.

As HTLV-1-associated diseases are more frequently observed after the fifth decade of life, osteoarthritis is expected to occur in infected individuals, which can represent a confounding factor in the establishment of criteria for HTLV-1 associated arthropathy. Here we compared clinical manifestations and radiographic findings between HTLV-1-infected individuals with a history of joint pain for at least one year and patients with osteoarthritis. As inflamed joint is not a common manifestation in HTLV-1 infected subjects, we did not considerer to include as control patients with systemic rheumatic disease. Our findings indicate that while a polyarticular and symmetrical pattern of joint involvement was observed in HTLV-1-infected patients, individuals with osteoarthritis presented predominantly an oligoarticular and asymmetrical involvement. While osteophytes and joint space narrowing was more frequent in patients with osteoarthritis, enthesophytes were the main radiographic finding in individuals with HTLV-1.

The incidence of osteoarthritis and most systemic rheumatic diseases is greater in females [24] and in our series joint pain was more frequent in females of both groups. Accordingly, we cannot exclude that the greater number of women studied herein may be related to women's greater interest in seeking medical care [25]. HTLV-1-associated diseases, including arthropathy, are more frequently observed in patients with HAM/TSP than in carriers [26]. However, in our study, the observed rates of chronic joint pain were similar in HTLV-1 carriers and in patients with HAM.

Radiographic findings in osteoarthritis are mainly characterized by the presence of osteophytes, joint space narrowing, bone cysts and subchondral bone sclerosis [19]. Accordingly, we found a higher frequency of osteophytes and joint space narrowing in controls with osteoarthritis compared to HTLV-1 infected patients. Alternatively, enthesophytes were more common in the HTLV-1 group. Enthesophytes are calcification that result from an inflammatory process where a tendon, ligament, aponeurosis or joint capsule is inserted into bone structure. This is a very common finding in inflammatory seronegative arthropathies, which include ankylosing spondylitis and psoriatic arthritis [27]. Here in addition to the enthesophytes bursitis and osteitis were found in the MRI of ankles in HTLV-1 infected subjects. The most relevant finding in the HTLV-1 infected group was the high frequency of enthesophytes. They were documented in the hips, knees and ankles, but were more frequent in the ankles of HTLV-1 infected individuals than in the control group. Enthesophytes are also common in patients with osteoarthritis and they were observed in the control group. However, the occurrence of enthesopathy in the absence of osteophytes and joint space narrowing was only observed in HTLV-1-infected subjects.

The majority of the diseases associated to HTLV-1 are mainly observed in patients with definite HAM/TSP [5,6,7]. However, we have previously shown that HTLV-1 carriers present higher frequency of HTLV-1 associated diseases than seronegative controls. Here the pattern of joint pain, the main joint involved and the radiographic features were similar in HTLV-1 carriers and in patients with probable and definite HAM/TSP indicating that the pathogenesis of the HTLV-1 associated arthropathy may be different from the one related to neurologic diseases, but future studies are necessary to determine how the virus and or the abnormalities in the immune response observed in HTLV-1 infection may cause arthropathy.

One of the limitations of our study was the fact that more specific and more sensitive imaging techniques for the detection of joint involvement were not used. However, conventional radiography, besides being a cheap, rapid and easily accessible method, does contribute to characterization of joint pathologies in the patients studied. Importantly, alterations detected on conventional radiography continue to be part of the diagnostic criteria for osteoarthritis [17]. In rheumatoid arthritis, joint features detected in conventional radiography were part of the diagnostic criteria of the disease up to 2010 when due to the limitation to detect of early changes in the count it was list out [28]. In patients with rheumatoid arthritis a characteristic finding of the disease is joint destruction [29] and it is important to highlight that in the present study, joint destruction was not observed in any HTLV-1 infected patient with joint pain.

Understanding the existence and importance of structures adjacent and related to the enthesis that contribute to dissipate the forces of stress helps to better understand the processes that affect the entheses. The study of enthesis of the calcaneus tendon should be extended to adjacent fibrocartilage, bursa, fat and fascia [30]. While the calcaneus enthesis is avascular in its fibrocartilaginous region, vessel growth may occur in association with the tissue repair response. The synovium, present in the retrocalcaneal bursa, is vascularized and vulnerable to inflammation and the high levels of pro-inflammatory cytokines observed in HTLV-1 infection may explain a higher frequency of retrocalcaneal bursitis and osteitis in the HTLV- 1 infected subjects compared to patients with osteoarthritis.

Conclusion

Based on our study HTLV-1 associated arthropathy is characterized mainly by symmetric polyarticular pain and enthesophytes in the absence of joint space narrowing and osteophytes in the ankles.