Persistent arthralgia among Chikungunya patients and associated risk factors in Chennai, South IndiaV Ramachandran, P Kaur, K Kanagasabai, S Vadivoo, MV Murhekar
National Institute of Epidemiology (ICMR), R-127, Tamil Nadu Housing Board, Ayapakkam, Chennai, Tamil Nadu, India
Correspondence Address: Source of Support: The study was funded by National Institute of Epidemiology, Chennai., Conflict of Interest: None DOI: 10.4103/0022-3859.128795
Source of Support: The study was funded by National Institute of Epidemiology, Chennai., Conflict of Interest: None
Context: Chikungunya (CHIK) fever is viral disease characterized by joint pain for prolonged duration in various settings. However, there are no reports of long-term follow-up of the CHIK patients from India. Aims: We conducted a cohort study to describe the clinical manifestations, incidence of persistent arthralgia, and the associated risk factors among patients with CHIK identified during an outbreak in a suburb of Chennai, India. Materials and Methods: We conducted a retrospective cum prospective cohort study in Gowripet, Avadi, Chennai. We included all adult CHIK case patients identified during the outbreak. We conducted a nested case-control study to identify the risk factors for persistent arthralgia defined as a CHIK case experiencing arthralgia for more than 15 days from the date of onset of illness. We included all 81 patients and 81 randomly selected controls. Results: All 403 case patients had joint pain. Approximately 40% suffered joint pain for up to 1 month and 7% had it beyond 1 year. The most commonly affected types of joints were knee (96%), wrist (80%), and ankle (77%) joints. Regarding the number of types of joints affected, 36% had six types of joints, 23% had five types of joints, and 14% had three types of joints affected. The overall incidence of persistent arthralgia was 80%. High-grade fever, involvement of four or more types of joints, and joint swelling were significantly associated with persistent arthralgia. Conclusions: High prevalence of persistent arthralgia indicates the need for appropriate treatment strategies to reduce the severity and duration of joint pain.
Keywords: Arthralgia, Chikungunya, India
Chikungunya (CHIK) is a viral disease with symptoms of fever, skin rash, and severe arthralgia/arthritis, involving mainly the small joints of the hands, wrist, and ankles, and skin rash. Joint pain is severe, and could last from a few days to several months.  Many countries such as the Comoros, Seychelles, Mauritius, and Reunion Islands experienced CHIK outbreaks in 2004-2005.  Fever epidemics with arthralgia/arthritis and rash similar to CHIK have been recorded as early as 1824 in India.  In India, CHIK fever was first reported in 1963 from Kolkata.  The last outbreak was reported in 1973 from Barsi in Maharashtra.  After nearly 32 years, it reappeared in an explosive manner in late 2005 with large outbreaks being reported from several Indian states. 
Joint pain is a characteristic clinical feature in viral illnesses caused by CHIK and many other alpha viruses.  There are few studies that have described evolution of joint pain among patients with CHIK and there are no reports of long-term follow-up among CHIK patients from India. ,,, and hence the present study was conducted. Our objectives were to describe the clinical manifestations, estimate the incidence of persistent arthralgia (PA), and determine the risk factors associated with PA among adult CHIK patients.
Human subject protection
This study was approved by the Scientific Advisory Committee of the National Institute of Epidemiology. This study was conducted in continuation of an emergency response to an outbreak. Written, informed consent was obtained from all participants.
A adult resident of Gowripet suffering from fever and joint pain between May and June 2006 and laboratory confirmation in the sub sample." 
Gowripet is located in Avadi, a suburb of Chennai, the fourth largest metropolitan city. Gowripet consists of 657 households and a population of 2649, of whom 57% ere females. There is scarcity of water, and water is stored in water storage containers, including plastic/earthen pots, plastic drums, and cement cisterns. The mean number of containers per household was nine. 
Study design, participants
During the outbreak investigation, all households in the outbreak affected area were enumerated and all the households were included in the survey. Patients were identified by house-to-house survey during the outbreak.  All patients were followed up until yhey had fully recovered from clinical symptoms. A nested case-control design was used to identify the risk factors for PA. We defined a case of PA as a CHIK case experiencing arthralgia beyond the acute phase of illness, i.e. more than 15 days from the date of onset of illness. Controls were CHIK patients who did not meet our case definition. Among 403 patients, only 81 patients did not have PA, and therefore, all were taken as controls. Among 322 cases with polyarthralgia, 81 were selected using random number table.
Data collection and analysis
We developed a semi-structured questionnaire and pilot tested the questionnaire. Study subjects were interviewed by trained field investigators using this questionnaire. Data quality checks were carried out in every 10 th case by trained field supervisors. Data were entered twice by two independent data entry operators.
Data was analyzed using Epi Info (3.52). Mean differences of the selected characteristics for CHIK patients with and without PA were compared. While testing the difference in the means of samples from two independent normal populations, we generally assume the variances to be equal. However, since unequal variances were encountered, we used modified t-test where a correction is designed to provide a valid t-test by correcting the number of degrees of freedom. We compared the medians using sample median test, and P value <0.05 was considered significant. We estimated the incidence of PA by including the total number of CHIK patients with PA in the numerator and the total population at risk (i.e. the total number of CHIK patients) in the denominator, expressed as a percent.
We did univariate analysis with PA as dependent variable and various socio-demographic and clinical manifestations as independent variables and calculated Crude Odds Ratios (COR). We performed multivariate analysis using logistic regression method and computed adjusted Odds Ratios (AOR) with 95% confidence Intervals.
Profile of study subjects
All 403 adult CHIK patients agreed to participate in the study and were included. Majority (90%) of the houses were pucca houses with no in-house tap facilities. Since water supply was infrequent (once in every 3-10 days), the residents were compelled to store water for their daily needs.
Males constituted 40%; mean age was 37.7 years (SD 14.3); median monthly family income was $66.5 (minimum $19, maximum $950). Majority [298 (74%)] were between 15 and 44 years of age. Among the patients, 148 (37%) had no education or education up to primary level, 154 (38%) were educated up to middle school, and the remaining had either high school or college level education.
Majority [318 (79%)] of the patients had high fever. Mean duration of fever was 7.3 days (SD 11.9). Approximately 65% complained of headache. Higher proportion of females reported headache compared to males (69% vs. 58%; P < 0.05). Nausea and vomiting were reported by 43% and 44% of the patients, respectively. About 11% of patients developed rash, with the proportion being higher among females (12.4%) as compared to males (9.3%). Median duration of rash was 5 days (minimum 1 day, maximum 30 days). One-third of the patients had rash lasting for 10-30 days.
Joint pain and types of joints affected
All cases had joint pain. Approximately 40% suffered joint pain for up to 1 month, 23.3% for 1-3 months, 16% for 3-6 months, 12.7% between 6 and 12 months, and 7% beyond 1 year. Approximately 0.74% (3/403) suffered joint pain for nearly 30 months. Mean duration of joint pain was 114.4 days (SD 157.9) and the median duration was 59 days (range 3-872 days). Since the means and medians were different, we have reported the medians here. The median duration of joint pain was 82 days (range 3-872 days) for patients above 35 years and was 31 days (range 3-653 days) for patients aged below/up to 35 years (P < 0.001). Similarly, the median duration of joint pain for males was 30 days (range 1-653 days) and for females was 62 days (range 1-865 days) (P < 0.001).
The type of joints affected included knee (96%), wrist (80%), ankle (77%), phalanges (77%), tarsals (72%), metatarsals (48%), shoulder (4%), elbow (2%), and hip (3.2%) joints, with no significant difference between males and females. Phalanges and tarsals had significantly higher involvement among patients above 35 years compared to those below 35 years (P < 0.05). Age had no effect on the affliction of other joints.
Regarding the number of types of joints affected, 36% had six types of joints, 23% had five types of joints, 14% had three types of joints, and only 6% had one type of joint affected. The mean and median number of types of joints affected were 4.6 (SD 1.6) and 5.0 (range 1-7 days), respectively. There were no significant differences with respect to age and sex with regard to the number of types of joints affected.
Nearly 46% of patients reported swelling of the joints with the proportion being significantly higher among females as compared to males (54% vs. 34%; P < 0.05). Similarly, higher proportion of patients above 35 years had joint swelling as compared to patients below 35 years (55% vs. 40%; P < 0.001).
PA and risk factors
The overall incidence of PA among CHIK patients was 80% (322/403; 95% CI: 75.8-83.6). The incidence among females was 83% (201/242; 95% CI: 77.9-87.4) and among males was 75% (121/161; 95% CI 68.1-81.4). We compared the various characteristics for CHIK patients with and without PA. Mean age, duration of fever, the number of joints affected, duration of joint swelling, and duration of joint pain were significantly higher for patients with PA compared to patients without PA [Table 1].
Among the risk factors, age above 35 years [2.0 (95% CI: 1.1-4.0)], fever for more than or equal to 4 days [3.6 (95% CI: 1.8-7.4)], high-grade fever [4.6 (95% CI: 2.1-10.5)], affliction of several small joints (ankles, wrist, phalanges, tarsals, and metatarsals), involvement of more than four types of joints [3.4 (95% CI: 1.7-6.8)], swelling in the joints [8.6 (95% CI: 3.9-19.2)], and presence of rash [4.0 (95% CI: 1.2-15.3)] were significantly associated with PA in univariate analysis (P < 0.05) [Table 2]. In multivariate analysis, high-grade fever [3.8 (95% CI: 1.6-9.0)], involvement of more than four types of joints [2.4 (95% CI: 1.1-5.1)], and joint swelling 6.0 (95% CI: 2.7-13.1) were significantly associated with PA [Table 3].
The present study showed PA in a large proportion of patients with involvement of multiple joints beyond acute illness in an urban South Indian population. Resurgence of CHIK in the past few years in several Asian countries has raised the concern over this long-term complication of the disease.
Several studies have estimated the prevalence of PA among CHIK fever patients at various time points during follow-up. ,,, A large proportion of patients had PA beyond 15 days in our study. Our findings are consistent with those of two studies conducted among travellers returning from the Islands in the Indian Ocean. In one study, 80% (38/47) had joint pain beyond 10 days and in another study, 69% (47/69) had joint pain beyond 2 months. , We observed persistent pain among 7% beyond 1 year, which is comparable to the observations from South Africa where 6% (at 36 months) had PA. Borgherini et al. followed up 88 patients for 18 months in Reunion Islands; among them, 63% had polyarticular arthralgia.  Wide variation in the prevalence of PA beyond 1 year might be due to the criteria used for inclusion of patients in the study. We included all the patients with the disease severity ranging from mild to severe in a geographically defined location following an outbreak, irrespective of the treatment and hospitalization status. Therefore, our prevalence reflects the estimates for the community included in the study. Our results might be strikingly different from those of studies that used patients from hospital settings or with any other inclusion criteria.
Involvement of multiple joints was one of the key characteristics of the disease in our study, with involvement of wrist, ankles, knees, and small joints of hands and feet. Similarly, in Reunion Islands, among 147 patients with PA followed up for 15 months, 76% had involvement of >4 joints at baseline with involvement of knees, ankles, wrist, and small joints of hands and feet.  Involvement of similar multiple joints has been reported in other studies. High-grade fever and prolonged swelling of joints were the risk factors that indicated the ongoing inflammatory response and manifested in the form of persistent joint pain. A study from Reunion Islands to identify the pathogenesis of PA at 12 months post-infection showed that patients with chronic pain had high viral loads in the acute phase indicating more severe disease. Persistence of CHIK in the joints might be responsible for chronic pain.  Identification of patients who are likely to have PA based on the clinical profile and understanding of the pathogenesis might play a key role in planning the long-term treatment.
Our community-based cohort study was unique in terms of follow-up of the patients affected during CHIK outbreak for a period of 30 months until all patients had reported that they were completely free of clinical symptoms. Our study findings provide evidence that CHIK fever outbreaks are unique in that in the prolonged chronic phase, they impair the physical functioning of patients. The limitation of our study was not performing laboratory tests to rule out other types of arthritis and also not carrying out radiological imaging of the joints.
Our study describes the high incidence of polyarticular PA in a CHIK outbreak setting in a suburban area in South India. This suggests the need for long-term supportive care and treatment. There is also a need to develop standard treatment guidelines that address the treatment needs of the patients in acute and chronic phases of the disease. Since the disease and its complications affect a large proportion of the population, development of a vaccine might be an effective intervention at the population level.
We thank the field staff from National Institute of Epidemiology (ICMR), the health authorities from Avadi Municipality, Chennai, and the study participants for their assistance, support, and cooperation. We thank Dr. MD Gupte for facilitating the conduct of the study. This study was internally funded by National Institute of Epidemiology of the Indian Council of Medical Research, New Delhi.
[Table 1], [Table 2], [Table 3]