Introduction
Chikungunya virus (CHIKV) infection causes a short period of acute fever with widespread rash, with limited impact on mor tality but with significant short and medium term sequelae in most of the individuals affected with joint involvement.1
According to PAHO, there were 351.334 suspicious cases reported in 2016 and 152.769 confirmed cases, with a global incidence rate of 50.51 case per every 1.000 inhabitants and 172 deaths throughout the Americas. 2 Some of the compli cations of the infection include chronic CHIKV arthropathy defined as the persistence of arthritis or arthralgia for more than 3 months, in a patient with clinical or laboratory confirmed infection. 3,4 Chronic CHIKV arthropathy is fre quently reported in the various series worldwide, with reports of up to 78% after 2 years of the acute infection episode. 5-7 This joint involvement affects people of any age; however, the prognosis tends to be worse in the elderly, leading to a negative impact on quality of life, on func tionality, and at the psychological level as measured with generic scales such as the 36-item Short -Form Health Survey Questionnaire (SF-36) and the General Health Questionnaire (GHQ-12). 7
Another aspect of the disease which is still being studied is the immune profile, since certain molecules that are pre dictors of chronicity have been identified, such as high TNF-a titers, interleukin (IL)-8, IL-6, and IL-12, 8-10 in the acute phase, and in the chronic phase the persistence of elevated IL-6, IL-17, IL-1B, IL-8 levels, monocyte chemoattractant protein-1, 10-12 suggesting interaction pathways between the initial innate immune response and the subsequent adaptative Th17 pathway as mediators of chronicity. All of these processes associated with the risk of causing erosion suggest the possibility of measuring the rheumatoid factor and the anti-citrullinated peptide antibodies as potential severity biomarkers.
In Colombia, the epidemics peak was identified between 2014 and 2015, reporting 19,566 accumulated cases by the end of13 and 1,110 cases on epidemiological week 50 in 2017, 14 indi cating a large volume of the population at risk of developing CHIKV arthropathy and its functional sequelae. It is impor tant to describe the clinical characteristics of the disease in our population and to identify prognostic predictors in order to be able to design therapeutic guidelines not only in terms of pharmacology, but also in terms of rehabilitation and social support. The objective of our study was to describe the clinical characteristics, the presence of rheumatoid factor seropositivity, anti-citrullinated peptide antibodies, and the levels of IL-6 and IL-17 as potential markers for chronicity. The functional disability and the impact on quality of life were also measured using generic scales in a sample of Colombian patients with CHIKV-related chronic arthropathy.
Materials and methods
This is and observational, cross-sectonal trial with a popula tion sample from San Juan de Nepomuceno City and municipal capital in Bolívar (Colombia), during the months of June and July, 2015. The inclusion criteria were 18 years or older, consis tent with the definition of clinically suspicious or confirmed cases for CHIKV infection and the presence of joint symp toms for more than 3 months; the exclusion criteria were a prior diagnosis or clinical suspicion of inflammatory or non-inflammatory underlaying osteoarticular disease and a suspicion or a clinical or laboratory diagnosis of infection from other arthritis-associated viruses. The sample com prised patients captured from hospital and community cases meeting the inclusion criteria and accepting to participate in the study and signing the informed consent. A venous blood sample was collected from the participants which was stored at -70 °C for transportation and further biochemical analysis. After collecting the blood sample, each patient was admin istered the HAQ-DI, SF-36, and DAS28 scales (osteoarticular examination conducted by a rheumatology resident), based on their standardization and validation as tools for assessing the severity of functional impairment, quality of life impact, and joint activity, respectively. A survey was conducted to col lect clinical and sociodemographic information. The samples were processed at a reference laboratory in Bogotá (Colombia) for the determination of C-reactive protein levels (PCR: turbid ity, Spin react turbilatex, SPINREACT), rheumatoid factor (RF: turbidity, Spin react turbilatex, SPINREACT), anti-citrullinated peptide antibodies (QUANTA LITE CCP3 IgG EIA, INOVA), IgG Chikungunya Virus (Human Anti-Chikungunya Virus IgG ELISA Kit, Abcam), IL-6 (Thermo Scientific™ Pierce™ Human IL-6 ELISA Kits, Thermo Fisher Scientific) and IL-17 (Thermo Scientific™ Pierce™ Human IL-17 ELISA Kits, Thermo Fisher Scientific). The protocol was submitted for approval of the Ethics Committee of the School of Medicine of Universidad Nacional de Colombia. All patients signed their informed consent for admission to the trial. The trial meets the requirements for research in humans, pursuant to Resolution 8430 of 1993, of the Colombian Ministry of Health, which is consistent with the rules and standards on human research ethics of the Declaration of Helsinki in 1964.
The following definitions were considered4
The clinic is suspicious: patient with fever, arthralgia or acute onset arthritis and rash; symptoms that cannot be explained by other medical conditions and original from a municipality which has not been declared outbreak zone.
Case confirmed based on the clinic: patient with fever, arthralgias or acute onset arthritis and rash; symptoms that cannot be explained by other medical conditions and original from a municipality which has been declared outbreak zone.
Case confirmed by the laboratory: suspicious case with any of the following laboratory tests positive for CHIKV: viral isolate, PCR-TR, IgM anti-chikungunya antibodies or 4-fold increase in the titer of IgG specific antibodies for CHIKV in matched samples collected 15 days apart.
Statistical analysis
All of the information from each variable collected via the data collection forms was coded and transcribed to a database in Excel 2010 (Microsoft office for Windows 10) for subsequent export and analysis using STATA 13 (Statacorp. 1985). The information obtained was reviewed in order to correct any information errors, avoid duplications, entry errors, or out liers. Statistical tools were used to describe the quantitative variables summarizing the central tendency measurements in accordance with the statistical distribution. In the case of qualitative variables, absolute sequences and percentages were used. The data shall be presented using tables and charts.
Results
Initially, 1850 patients were identified with a clinical presen tation of acute onset fever and arthralgias, of which 102 cases were clinically confirmed as CHIKV and were associated with arthropathy of more than 3 months. The clinical information, informed consent, and blood samples were collected for anal ysis in 94 patients. 72 patients (76%) were females. The mean group age was 57 ± 14.9 years, with a mean duration of symp toms of 278 ±87.8 days. By definition, 100% presented with symptoms extending beyond 90 days, with a minimum dura tion of symptoms of 91 days and a maximum of 365 days. 75.5% of the affected population came from a low socioeco nomic group and 45% of the total population had no schooling whatsoever (Tables 1 and 2).
95% of the cases presented with fever, skin rash, myal gia, head ache, fatigue, abdominal pain and nausea/emesis, low back pain in 63.4% and conjunctivitis and adenopathy were identified in less than 50%. Synovitis was identified in 29.8% of the cases at the time of the assessment, with a mean of 2.21 ± 1.26 swollen joints. The joints distribution is illus trated in Figs. 1 and 2, showing a predominantly symmetrical presentation, involving large and small joints of the upper extremities, knees and ankles.
According to DAS28 activity scale, 40.4% presented with low activity (DAS28 < 3.2), 55.3% had moderate activity (DAS28 3.2- ≥ 5.1), and only 4.2% (4 patients) had a high activity (DAS28 > 5.1). According to the HAQ-DI functional scale, 44.7% had mild involvement (score less than 1), 45.7% moderate (score between 1 and <2), and 9.5% severe (score of 2 or more). The generic SF-36 quality of life scale showed more involvement of the physical, emotional and pain components (percentages below 50%); however, all the domains of this test were affected and none scored above 65% (Table 3 and Figs. 3-5).
Immunologically, 100% of the participants had detectable IgG levels for CHIKV. RF positivity was identified in one patient (1.06%) and none of them were anti-CCP positive. The mean PCR levels were 4.42 ± 7.10 mg/l. Of the total num ber of patients, 64.9% presented detectable IL-6 levels, with a mean value of 4.51 ± 6.09 pg/ml in the total population and 7.45% of the total population had detectable IL-17 lev els, with a mean value of 1.51 ± 9.39 pg/ml. In the group of patients with arthritis, the IL-6 levels were 7.93 ± 6.16 pg/ml and in the group with no evidence of arthritis, the IL-6 levels were 6.43 ±6.47 pg/ml. The mean IL-17 levels in the arthritis group were 10 ± 13.57 pg/ml and in the non-arthritis group was 30 ±38.44 pg/ml. In the discrimination of IL-6 values accord ing to the degree of activity of the disease staged using DAS28, in the mild range of activity (<3.2) a mean IL-6 value of 4.14±5.93pg/ml was identified; in the moderate range value (3.2-5.1) the mean IL-6 was 4.61 ±6.40 pg/ml, and in the high activity (>5.1) was 6.82 ±3.48 pg/ml. In the case of IL-17, since 7 patients showed detectable levels (6 of them in the moder ate activity group, one in the high activity group and none in the mild activity group), the moderate and high activity groups were analyzed together, resulting in mean IL-17 values of 20.28 ± 30.39 pg/ml (Tables 3 and 4).
Discussion
The variability in clinical behavior of patients that develop chronic joint manifestations due to CHIKV is remarkable; some studies such as Chopra et al. 15 in India in 2012, reported just 4% of persistence of symptoms after one year; however, others report up to 60-75%, in accordance with French cohorts in Reunion island in 2013. 16,17 Hence there is a significant heterogeneity in joint involvement based on geography and probably the phenotype of the disease. The design of this study did not allow for any conclusions with regards to the rate of persistence of joint symptoms; however, it does identify some clinical characteristics of CHIKV-associated arthropathy in the Latin population. In our setting, the clinical symptoms per sisted for around 9 months following the acute presentation; approximately 2/3 of the population are females, consistent with the reports from other authors, 5,16,18,19 suggesting some type of susceptibility in women. The population identified was mostly from low socioeconomic sectors, the majority with lit tle or no schooling, which possibly reflects some selection bias. However, we should not rule out the fact that this population may have a higher risk of developing the disease because of poor knowledge about management of drinking water storage, due to the association between this variable and the vector replication. 14
It should be highlighted that the predominant joint involve ment was symmetrical with synovitis - around 30% of the cases - similar to the findings reported in other case series that also report between 20 and 65% of synovitis, 5,20,21 and the pres ence of inflammatory type of pain, regardless of the presence of joint inflammation in up to 70% of the patients affected. 6 The international literature reports variable frequencies of RF and anti-CCP positivity, reporting between 12 and 43% of RF positivity in India, 22 and between 30 and 50% for RF/anti-CCP according to French series, 19,23 probably indicating a more severe clinical profile in some of these groups, and with a higher risk of erosion. Prior studies in the Colombian popu lation report 4.2%24 seropositivity for these antibodies, which is similar to the findings by Manimunda et al. 20 also in India and Schilte et al. 17 in France; this may be a reflection of dif ferent selection criteria in the various trials. Some may have selected a population with more severe inflammatory involve ment or patients with pre-existing undiagnosed inflammatory arthropathy.
The consistency of the seropositivity data in our popula tion may reflect a less aggressive phenotype with less risk of progression to erosive disease in the long term. Further longitudinal and follow-up studies are needed to clarify this hypothesis. Similarly, the identification of a low prevalence of IL-17 elevation supports the idea of a less severe clinical profile and possibly an improved prognosis in the long term, or the likely selection of patients in a relatively early stage (the first year of the disease) in which the Th17 line differentiation is not yet identifiable, as is the case in other types of inflammatory arthritis.
In terms of the cytokines profile, the percentage of patients with detectable IL-6 was significant in our population - 65% , similar to the data reported by Jaller et al. 24 also in the Colombian population, with 95% positivity for this marker. Moreover, the IL-17 levels in our studies were only detectable in 7.4%. This highlights the significant role of IL-6 in the patho physiological mechanism for the joint inflammatory process, without ruling out other systemic effects. One of the limi tations in our study was the cross-sectional assessment of patients, which limited the possibility to do a long term eval uation of the cytokines profile and their evolution over time.
CHIKV-associated arthropathy has shown a significant impact on the functional, emotional and quality of life aspects of the different populations affected, 7,19,20 with an estimated economic impact in the Reunion island of 34 million Euros per year, between 2005 and 2006. 17 Our population evidenced the same pattern reported with significant compromise of health-related quality of life, mainly in terms of pain, physical and emotional performance, but in general of all the aspects assessed with the SF-36 scale. In terms of the functional capacity assessment, the generic HAQ-DI25 identified in our population a moderate functional involvement, in around fifty percent of the population and a severe compromise defined as an HAQ-DI score between 2 and 3 in almost 10% of the patients. This indicates that the level of functional limitation in this population, most of them in working age, generated a signifi cant impact from the labor and social point of view. Since there are no long term studies that clearly typify the evolution over time, one may suspect that the functional and quality of life compromise in these patients could be affected, and so it is necessary to conduct long term prospective trials assessing the impact of the disease the labor, social and psychological realms.
The limitations of our study were its cross-sectional nature that prevented follow-up over time to conduct prospective assessments of erosion risk markers (RF and anti-CCP) and the fact that our population exhibited a relatively early clini cal evolution as compared to other studies involving patients with 1 and 3 years of evolution of symptoms.
In conclusion, our study evidenced the significant func tional and quality of life involvement of the affected patients, with almost one third presenting with arthritis in the first 9 months following the acute infection and more than 60% with detectable IL-6 levels as a potential immune marker for chronicity, but with a low seropositivity for rheumatoid fac tor and anti-CCP. Further prospective studies are required to follow-up these populations in order to monitor their clin ical and functional behavior, as well as the serum markers for severity. Due to the significant quality of life impact, it is important to adopt therapeutic measures with medications and social and functional interventions.