Search

Targeted treatment of rheumatoid arthritis (RA) includes biological DMARDs (bDMARDs) and JAK inhibitors (JAKi). These agents are recommended at the same level on the basis of their efficacy and safety data. However, no local evidence of the impact of RA treatment regimens on total budget spending is available to date. This study aimed to explore the budget impact of different sequential targeted treatments in Thai patients with RA who failed at least three conventional synthetic DMARDs. We used the adapted model to evaluate the budget impact of adding tofacitinib in different order to RA targeted treatment regimens. The Thai RA population eligible for treatment was assessed on the basis of local prevalence and experts’ opinion. Cost-impact analysis was evaluated for the treatment sequences of four different lines of targeted therapies using inputs like clinical efficacy, safety, and costs. The model used a decision tree structure with treatment nodes corresponding to treatment response outcomes for a cohort of patients. The comparisons included five bDMARDs [etanercept (ETN), infliximab (IFX), golimumab (GOL), rituximab (RTX), tocilizumab (TCZ) intravenous formulation], two JAKi [tofacitinib (TOF) and baricitinib (BAR)], and two IFX biosimilars (PF-06438179/GP1111 and CT-P13). A total of 80 treatment sequences within each containing four sequential first-, second-, third-, and fourth-line options were generated. The findings of the base case scenario indicated the treatment sequence with RTX as first-line, followed by IFX biosimilar (PF-06438179/GP1111), TOF, and TCZ, respectively, produced the lowest budget impact of US $693.54 million. Sensitivity analyses confirmed the robustness of our findings. The order of targeted therapy starting with RTX, then IFX biosimilar, TOF, and finally TCZ incurred the lowest budget impact over a 5-year time horizon for treating moderate to severe RA. Our findings may help payers and policy makers consider appropriate budget allocation on chronic non-communicable diseases, especially RA.

Resume Objectif Etudier le lien entre le syndrome metabolique (SM) et l’activite de la maladie chez les patients atteints de polyarthrite rhumatoide (PR). Methodes Le registre des polyarthrites rhumatoides de l’hopital Siriraj est une etude de cohorte prospective debutee en mai 2011. Un total de 267 patients disposant de donnees completes en fevrier 2015 a ete inclus dans ces analyses. Toutes les donnees cliniques et biologiques relatives a l’activite de la maladie, le statut fonctionnel et les parametres du SM selon le National Cholesterol Education Program Adult Treatment Panel III (NCEP ATP III) ont ete collectees. Des analyses univariee et multivariee selon la methode pas a pas descendante ont ete menees pour identifier les facteurs associes au SM. Resultats La majorite des patients (88 %) etaient des femmes, d’âge moyen (±ecart-type) 59 ans (± 11,1). Un SM etait retrouve chez 43 (16 %) des patients. La proportion de patients en remission selon le DAS28 (moyenne ajustee sur le temps du DAS28 Conclusion Le taux de remission est significativement plus bas chez les patients PR atteints du SM. L’activite de la PR, l’indice de masse corporelle et le niveau d’education sont associes au SM chez les patients atteints de PR.

Background:Rheumatoid arthritis (RA) and spondyloarthropathies (SpA) are chronic rheumatic diseases that can progress to disability if left uncontrolled. Biologic therapies can induce remission in patients with inadequate response or intolerance to conventional synthetic disease-modifying antirheumatic drugs (csDMARDs). Due to the safety and economic concern, updated treatment recommendations now consider biologics discontinuation in patients with persistent remission. However, available data are still limited.Objectives:To evaluate the biologics effectivenss, drug survival, rate of discontinuation in real-life practice and to identify predictors of biologics discontinuation due to remission in RA and SpA.Methods:A total of 87 patients with RA and 49 patients with SpA, starting biologics between January 2005 and October 2020, were recruited from the Rheumatic Disease Prior Authorization (RDPA) registry of Siriraj hospital. Baseline data were recorded. Time-dependent rates in achieving remission were calculated. Cumulative probability of biologics discontinuation and predictive factors of drug discontinuation due to remission were analyzed.Results:The biologics used in RA patients were Etanercept (49.4%), Rituximab (44.8%), and Infliximab (5.7%). The 1- and 5- year remission rates were 8.5% and 23.9%, respectively. Drug survival rates were 33.6% for Etanercept, 40.7% for Rituximab, and 36% for Infliximab at five years. The main reasons for drug discontinuation were disease remission (32.1%), inadequate response (32.1%), and side effects (15.1%). Rituximab had the highest discontinuation rate due to remission (56.4%). There was no predictive factor for biologics discontinuation due to remission.SpA patients were treated with Etanercept (83.6%) and Infliximab (16.4%). The 1- and 5- year remission rates were 67.4% and 66.7%, respectively. Drug survival rates were 50.3% for Etanercept and 0% for Infliximab at five years. Biologics withdrawal due to side effects occurred in 16%, inadequate response in 10%, and disease remission in 10% of patients. The predictive factor for biologics discontinuation due to remission was disease duration < 5 years (Hazard ratio 6.92, 95%CI 1.10, 43.29)Conclusion:Biologic therapies are effective in patients with active RA and SpA despite csDMARDs treatment. The highest drug survival rates are Rituximab in RA and Etanercept in SpA. Infliximab has the shortest drug retention rate in SpA. Drug discontinuation due to remission is more successful in RA patients, particularly in the Rituximab group. The best predictor for biologics discontinuation due to remission in SpA is the disease duration of fewer than five years.Table 1.Effectiveness, Drug Survival and Causes of Biologic Discontinuation in RA and SpARA % (87 cases)BiologicsTotal % (N)5-year Remission rate %5-year Drug survival rate %Discontinuation % (N 53 cases)Total %(53)RemissionInadequate responseSide effectNon-adherence/ ReferDeathEtanercept49.4 (43)23.933.650.9(27)14.8(4)44.4(12)22.2(6)11.1(3)7.4(2)*Infliximab5.7 (5)365.7 (3)0(0)33.3(1)33.3(1)33.3(1)(0)Rituximab44.8 (39)40.743.4 (23)56.5(13)17.4(4)4.3(1)13.0(3)8.7(2)SpA %(49 cases)BiologicsTotal % (N)5-year Remission rate %Drug survival rate%Discontinuation % (N 23 cases)Total % (23)RemissionInadequate responseSide effectNon-adherence/ ReferDeathEtanercept83.6 (41)66.750.378.3(18)22.2(4)22.2(4)33.3(6)22.2(4)0(0)Infliximab16.4 (8)021.7 (5)20(1)20(1)40(2)20(1)0(0)*Cause of death: sudden cardiac death (1), intraabdominal infection (1)Cause of death: pneumonia (1), Lung cancer (1)Disclosure of Interests:None declared

Objective To investigate the association between metabolic syndrome (MS) and disease activity in patients with rheumatoid arthritis (RA). Methods Siriraj Rheumatoid Arthritis registry is a prospective cohort study establishing since May 2011. A total of 267 patients who had complete data in February 2015 were included in these analyses. All clinical and laboratory data related to disease activity, functional status, and parameters of MS according to the 2001 National Cholesterol Education Program Adult Treatment Panel III (NCEP ATP III) were collected. Univariate and backward stepwise multivariate analyses were performed to identify factors associated with MS. Results Most (88%) were female with the mean age ± standard deviation of 59 ± 11.1 years old. MS was found in 43 patients (16%). Patients with MS had a significantly lower proportion of patients with remission (time-adjusted mean of disease activity score 28 or DAS28 P = 0.02). Multiple logistic regression analysis identified 3 independent factors associated with MS including body mass index [OR 1.2, 95% CI 1.1 to 1.3], educational level ≤ 12 years [OR 5.92, 95% CI 1.47 to 23.83], and disease remission [OR 0.11, 95% CI 0.01 to 0.93]. This model correctly predicted 84% of cases. Conclusion Remission rate is significantly lower in RA patients with MS. Disease activity of RA, body mass index, and educational level are associated with metabolic syndrome in patients with RA.

Objective This study aimed to identify factors associated with EuroQoL-5 Dimensions, 5 Levels and Patient Acceptable Symptom State (PASS) and health utility (HU) in Thai patients with ankylosing spondylitis (AS). Methods This was a cross-sectional study of consecutive AS patients visiting Siriraj Hospital between May 31, 2012, and March 31, 2016. Demographic data and outcomes related to HU (Thai version of EuroQoL-5 Dimensions, 5 Levels), disease activity (Bath Ankylosing Spondylitis Disease Activity Index, Ankylosing Spondylitis Disease Activity Score-erythrocyte sedimentation rate or Ankylosing Spondylitis Disease Activity Score-C-reactive protein, number of tender and swollen joints, and enthesitis), and functional status (Bath Ankylosing Spondylitis Functional Index and Health Assessment Questionnaire) were collected. Regression analysis was used to explore factors associated with each EuroQOL-5 Dimensions (EQ-5D) domain, HU, and PASS. Results Among 119 AS patients, the mean age was 40.4 years; 61.3% were male. The mean EQ-5D was 0.75. In univariate analysis, lower disease activity and less impaired function were significantly associated with higher HU and not to mild problems in each EQ-5D domain. In multivariate regression analysis, Bath Ankylosing Spondylitis Disease Activity Index, C-reactive protein, and Health Assessment Questionnaire adjusting for age explained 77.4% of the HU variance. Patients answering yes to PASS were significantly older, had higher HU, and lower disease activity compared with those answering no to PASS. Usual activity and pain problems were importantly related to PASS after adjusting for other domains and age. Conclusions Disease activity and functional status in AS patients were significant factors related to HU and PASS. To improve quality of life, treatment goals should be achieving remission, improving function, and controlling pain.

The objective of this study was to investigate the association of rheumatoid factor (RF) and anti-citrullinated peptide antibodies (ACPA) status with disease progression and treatment outcomes in patients with rheumatoid arthritis (RA). A total of 276 adult patients who fulfilled the American College of Rheumatology 1987 classification criteria for RA were recruited from the Rheumatology clinic, Siriraj Hospital, from January 2011 to December 2012. Demographic, clinical, and laboratory data were collected at baseline and every 3 months up to 1 year of follow-up. RF and ACPA were measured at baseline. Radiography of the hands and feet was performed at baseline and 1 year. Patients with RF+/ACPA+ had significantly more severe disease activity and impaired functional status than those who had RF−/ACPA−. Although they received more aggressive treatment with methotrexate and combination of non-biologic, disease-modifying antirheumatic drug than other groups, fewer patients in this group achieved remission at 1 year of follow-up, especially when compared to RF−/ACPA− group (12 vs. 18 %). For radiographic erosion, patients with the presence of either RF or ACPA had a higher proportion of hand erosion than seronegative patients at baseline (77, 73, 83, and 32 %, p

Aim In June 2015, the Thai Rheumatism Association (TRA) approved an update of its recommendation for the use of biologic disease-modifying anti-rheumatic drugs (bDMARDs) and targeted synthetic (tsDMARD) in the treatment of rheumatoid arthritis (RA) to cover those currently available in Thailand (etanercept, infliximab, golimumab, rituximab, tocilizumab, abatacept and tofacitinib). Method A search of the literature was performed between January 2000 and June 2015. Existing RA recommendations, in relation to the use of bDMARDs and tsDMARD, were identified and evaluated by the AGREE II instrument prior to their use as a 'guide' for developing this TRA recommendation. An additional literature search was performed in order to answer specific clinical questions that could not be found in existing guidelines. Result Thirteen recommendations were developed. They covered the use of RA classification criteria, the aim of RA treatment, when to initiate bDMARDs/tsDMARD or taper or switch them to other medications, as well as monitoring these drugs during their use. In addition, specific issues including their use and vaccination, malignancies, pregnancy and lactation, and perioperative period also were addressed. Public hearings were performed at the annual meeting of the TRA and of the Royal College of Physicians of Thailand. The recommendations were distributed to other professional associations related to RA management, as well as government sectors associated with the reimbursement policy, prior to development of the final version. Conclusion These recommendations will help Thai rheumatologists prescribe bDMARDs and tsDMARD more appropriately when treating RA patients.

Aim To determine the prevalence and to identify the risk factors of chloroquine maculopathy (CM), and to evaluate the association of plasma chloroquine (CQ) and desethylchloroquine (DCQ) levels and CM. Methods Rheumatoid arthritis (RA) patients who had taken CQ for at least 6 months and stable CQ dosage for at least 2 months were included. CM was diagnosed by dilated ocular examination and automated visual field. Plasma CQ and DCQ levels were determined by liquid chromatography tandem mass spectrometry method. Logistic regression was used to explore risk factors associated with CM. Results One hundred and ninety-three patients were included with median CQ duration (range) of 50.2 months (6.0–269.8) and cumulative dose of 137.4 g (16.4–1226.5). The prevalence of CM was 13.5%. Factors associated with CM identified from univariate analysis were age > 60 years, and creatinine clearance with odds ratio (OR) (95%CI) of 5.79 (2.42, 13.84), and 0.98 (0.96, 1.00). In multivariate analysis, older age, usage > 5 years, and current dose from 2.5 mg/kg ideal body weight [IBW]/day were the factors significantly associated with CM with OR of 5.89 (2.38, 14.57), 2.94 (1.10, 7.83), and 3.32 (1.04, 10.60), respectively, while plasma CQ and DCQ showed no association with CM. Conclusions The prevalence of CM was 13.5% among RA patients taking CQ for at least 6 months. Age > 60 years, duration of CQ usage > 5 years and current CQ dose ≥2.5 mg/kg IBW/day were the risk factors for CM. The plasma CQ or DCQ levels demonstrated no correlation in developing CM.

We report a newly recognized bone disorder consisting of polyostotic expansile osteolysis affecting long bones and iliac bones; hyperostosis of the skull, thoracic cage, and medial portion of both clavicles; pectus carinatum; gigantiform synovial masses of the elbows and knees; atrial septal defect; cardiomegaly; unilateral cryptorchidism; and mental deficiency. Affected bones can be grouped into four general types of skeletal pathology: (1) expansile osteolysis, (2) osteolysis without expansion, (3) expansion without osteolysis, and (4) hyperostosis. Some bones remained unaffected. We have named the condition "polyosteolysis/hyperostosis syndrome." It is clearly at variance with any previously reported bone disorder, including familial expansile osteolysis, juvenile Paget disease, and McCune-Albright syndrome (and polyostotic fibrous dysplasia). Because our patient shared some features in common with juvenile Paget disease, we thought that mutational analysis of TNFRSF11B was indicated, even though our patient had some manifestations not found in juvenile Paget disease. Direct sequencing failed to identify a TNFRSF11B mutation. Because the parents of our propositus were first cousins suggests that polyosteolysis/hyperostosis syndrome may possibly have autosomal recessive inheritance.

Background SLE mostly affects the working-age population. The disease burden has a significant impact to the quality of life. Objectives To assess the HRQoL in SLE patients by using the EQ-5D and the LupusQoL and to evaluate their correlation. Methods 248 patients were recruited. Demographic and clinical data including the SLEDAI were collected. We also evaluated the patients9 preference in terms of the language, the precision of questions, and time consume. The correlation was examined by Pearson9s correlation. Results 94.6% of patients were female. Mean age was 37.5±11.9 with the median disease duration of 5.1 years. The mean SLEDAI score was 3.6±0.4. The mean of EQ-5D health value and VAS scores were 0.65±0.20 and 78.1±20.2. Most patients reported that SLE had mild effect to their health status. 28.1 and 18.3% of patients reported moderate to severe problems in the mental components of pain/discomfort and anxiety/depression. The mean LupusQoL score was 78.9±16.3. Both EQ-5D health value and VAS were significantly correlated with all LupusQOL domains, strongly correlated with physical health and pain but poorly correlated with planning and body image. The patients preferred to use the EQ-5D over LupusQoL. Conclusions HRQoL in most SLE patients was generally good. The EQ-5D score was significantly correlated with the LupusQoL score particularly in the domains of physical health and pain. The EQ-5D is a simple and more convenient tool to assess the general health and can be used as a brief evaluation of lupus-specific HRQoL in daily practice. Disclosure of Interest None declared

Rheumatoid arthritis has manifestations in various organs including ophthalmic involvement. The present study evaluates prevalence of dry eye and secondary Sjogren's syndrome using salivary scintigraphy which has not been used in previous reports.To evaluate the prevalence of secondary Sjogren's syndrome in patients with rheumatoid arthritis, including clinical characteristics and dry eye, compared with non-Sjogren's syndrome.Descriptive cross sectional studySixty-one patients with rheumatoid arthritis were recruited at Siriraj Hospital during March 2009-September 2010 and filled in the questionnaires about dry eye for Ocular Surface Disease Index (OSDI) with a history taking of associated diseases, medications, duration of symptoms of dry eyes and dry mouth. The Schirmer I test without anesthesia, tear break-up time, rose bengal staining score, severity of keratitis and salivary scintigraphy were measured and analyzed.Prevalence of secondary Sjogren's syndrome and dry eye were 22.2% (95% CI 15.4 to 30.9) and 46.7% (95% CI 38.0 to 55.6), respectively. Dry eye interpreted from OSDI, Schirmer 1 test, tear break-up time and rose bengal staining was 16.4%, 46.7%, 82% and 3.3% respectively. Fifty-two percent of patients had a history of dry eye and dry mouth with mean duration 27.4 and 29.8 months, respectively. Superficial punctate keratitis and abnormal salivary scintigraphy were found in 58.2% and 77.8%. Duration of rheumatoid arthritis, erythrocyte sedimentation rate were not correlated with secondary Sjogren's syndrome. Dry eye from OSDI with secondary Sjogren's syndrome (33.3%) compared with non-Sjogren's syndrome (9.5%) was significant difference (p = 0.008). Adjusted odds ratio for secondary Sjogren's syndrome in OSDIL score25 was 13.8 (95% CI 2.6 to 73.8, p = 0.002) compared to OSDI score25.Awareness and detection of dry eye syndrome and secondary Sjogren's syndrome in rheumatoid arthritis was crucial for evaluation of their severity and proper management.

To determine the sensitivity, specificity, and agreement of Amsler grid (AG) in detecting antimalarial induced maculopathy (AM) compared to combination of dilated ocular examination and Humphrey visual field (HVF) 10-2 and to estimate the risk factors of AM.20 patients with AM and 20 patients with no AM were included between October 1, 2010 and June 30, 2011 in Siriraj Hospital, Thailand. Sensitivity, specificity and unweighted kappa were used to determine accuracy of AG. Logistic regression was used to explore risk factors associated with AM.The sensitivity and specificity of AG were 40% and 100%, respectively. The agreement of AG and HVF 10-2 was poor with kappa of 0.4. Creatinine clearance60 ml/min/1.73 m2 and bilateral macular abnormality were associated with AM at the same odds ratio of 8.9 (95% confidence interval: 1.3 to 61.2).AG may be used as an additional test between ocular evaluation visits providing the availability and low cost. Patients with renal impairment and bilateral macular abnormality are associated with AM.

To assess the predictive value of in-training evaluation for determining future success in the internal medicine board certifying examination.Ninety-seven internal medicine residents from Faculty of Medicine Siriraj Hospital who undertake the Thai Board examination during the academic year 2006-2008 were enrolled. Correlation between the scores during internal medicine rotation and final scores in board examination were then examined.Significant positive linear correlation was found between scores from both written and clinical parts of board certifying examination and scores from the first-year summative written and clinical examinations and also the second-year formative written examination (r = 0.43-0.68, p0.001). Monthly evaluation by attending staffs was less well correlated (r = 0.29-0.36) and the evaluation by nurses or medical students demonstrated inverse relationship (r = -0.2, p = 0.27 and r = -0.13, p = 0.48).Some methods of in-training evaluation can predict successful outcome of board certifying examination. Multisource assessments cannot well extrapolate some aspects of professional competences and qualities.

To compare the efficacy of heparin and EDTA and determine the impact of time delays in stabilizing leukocyte counts in synovial fluid.33 specimens were collected in heparin-preserved and EDTA-preserved containers. Total cell count was performed manually at 1 hour and 24 hours. Correlation between cell counts from both preservatives and the leukocyte number at 1 hour and 24 hours were analyzed by means of agreement measurement.There were good correlations between the leukocyte numbers from the specimens preserved by heparin and EDTA (ICC = 0.889, r = 0.879, P0.0001 at 1 hour and ICC = 0.822, r = 0.693, p0.0001 at 24 hour). At 24 hours, total cell counts from EDTA-preserved samples were comparable to those obtained at 1 hour (ICC = 0.985, r = 0. 986, p0. 0001) and were not different from those of the heparinized samples (ICC = 0.833, r = 0. 751, p0.0001) but the ICC value was higher.EDTA was as effective as heparin for preservation of synovial fluid. Therefore, it can be used routinely as a preservative of synovial fluid.

To determine the prevalence of rheumatic diseases in HIV-infected patients at Siriraj Hospital, Thailand.178 patients who attended the HIV-Clinic at Siriraj Hospital between November 2002 and February 2003 were examined for the presence of rheumatic diseases. Diagnosis of HIV infection was performed by ELISA and confirmed by partial agglutination testing. HIV-infected patients were classified according to the Centers for Disease Control (CDC) 1993 revised classification system. Standard criteria were used to classify the rheumatic diseases.98 patients had rheumatic diseases. Seventy-seven patients were treated with antiretroviral drugs. Forty-nine patients had mechanical low back pain, twenty-four patients had arthralgia, nineteen patients had plantar fasciitis, eighteen patients had nonspecific myalgia, thirteen patients had fibromyalgia, and eleven patients had others. Arthralgia was associated significantly with Quadricept muscle wasting (p = 0.00001). Nonspecific myalgia was more likely to be associated with female (p = 0. 018) and less likely with use of antiretroviral therapy (p = 0.031).Rheumatic diseases were commonly found in HIV-infected patients. Arthralgia associated with wasting Quadricep muscle. Nonspecific myalgia was predominant in female and without antiretroviral drug treatment.

Background Systemic lupus erythematosus (SLE) may vary in presentation and management approach in different geographies (1). A review of the literature describing the epidemiology of SLE in Asia-Pacific countries identified a dearth of information on the characteristics and perceived wide variability in management of lupus in the region (2). This study was designed to observe SLE patients and their care, from six countries in Asia-Pacific, under a common protocol, over a twelve month follow-up period. This abstract describes the baseline features of the cohort. Objectives The aim of the study is to describe patient characteristics, organ involvement, disease activity and severity, and management in a multi-national observational cohort. Methods Patients were invited to participate from lupus clinics in six countries (nine centres), between April and October 2012. Consecutive patients, 18 years or older, meeting at least four ACR criteria for diagnosis of SLE were eligible. Information describing organ system involvement, disease activity and severity, and management of the patients were recorded. Severe disease was defined as a) at least one of the major domains: cardiovascular, respiratory, renal or urological actively involved (biologically active eg proteinuria or symptomatic) at enrolment, and b) requiring more than 7.5mg/d of corticosteroids (prednisone equivalent dose) and/or immunosuppressant(s) and/or biological drugs. Results 553 patients (404, 92% women) were recruited. Mean age at SLE diagnosis was 28.8 (SD 11.4) years, and mean number of years since diagnosis was 9.2 (SD 6.7). For disease activity profile at entry into the study, the most common domain of active disease involvement was renal disease (17% of patients). Similarly, the domain most reported as being damaged (non-reversible change not related to active inflammation) since onset of lupus was renal (9%). 20% of the patients were classified as having severe disease, with Taiwan showing the highest proportion of severe disease (39%), and South Korea the lowest proportion (14%). Steroid use was reported in 81.6%, anti-malarials in 63.2%, immunosuppressants in 48.3% and a biological drug in one patient (0.2%). Conclusions We have provided a cross-sectional overview of the demographics, disease characteristics and therapies in a cohort of SLE patients seen at selected sites across 6 countries in Asia-Pacific. A high proportion of patients, at the time of enrolment, were managed with steroids, anti-malarials and immunosuppressants, with one patient receiving biological therapy. This sets the baseline for a one-year observational study utilizing a standard electronic case report form, in order to better understand the behavior and management practice of SLE in the region. References Danchenko et al. Epidemiology of systemic lupus erythematosus. Lupus 2006; 15: 308-18. Jakes RW et al. Systematic review of the epidemiology of systemic lupus erythematosus in Asia-Pacific. ACR 2012; 64: 159-168. Disclosure of Interest R. W. Jakes Shareholder of: GSK, Employee of: GSK, S. Navarra Grant/research support from: GSK, Speakers bureau: GSK, Roche, P. Jain Shareholder of: GSK, Employee of: GSK, W. Louthrenoo: None Declared, S.-C. Bae: None Declared, A. Hoi Grant/research support from: UCB, Eli Lilly, Suppremol, GSK, Consultant for: BMS, Speakers bureau: Abbott, BMS, MSD, Mundipharma, UCB, N. C. T. Kong Grant/research support from: GSK, A. Koolvisoot: None Declared, H.-Y. Lin: None Declared, P. Nash Grant/research support from: GSK, Consultant for: GSK, M.-Y. Weng: None Declared

Objectives To investigate the risk factors for chloroquine maculopathy (CM), and to explore factors associated with plasma chloroquine (CQ) and desethylchloroquine (DCQ) levels. Methods Two-hundred and thirty-three rheumatoid arthritis (RA) patients who had taken CQ at least 6 months with stable CQ dosage for at least 2 months were included between March 1, 2010 and September 30, 2011 at Siriraj Hospital, Thailand. CM was diagnosed by ophthalmologist utilizing dilated ocular examination and Humphrey visual field 10-2 white protocol. Plasma CQ and DCQ levels were determined by liquid chromatography tandem mass spectrometry method. Logistic regression models were applied to explore risk factors associated with CM. Results One hundred and ninety-three patients were included with median CQ duration (range) of 50.2 (6.0 – 269.8) months and cumulative dose of 137.4 (16.4-1226.5) gm. The prevalence of CM was 13.5%. One hundred and fifty seven plasma samples were available for determining the CQ and DCQ levels. Factors associated with developing CM identified from univariate analysis were age more than 60 years old, and poor creatinine clearance (CCl) with odds ratio (OR) (95%CI) of 5.79 (2.42,13.84), and 0.98 (0.96,1.00), respectively while current daily CQ dose from 2.5 mg/kg had marginal association, p=0.06. Plasma CQ and DCQ was not associated with CM with OR of 1.00 (1.00, 1.01) and 0.99 (0.98, 1.01), respectively. In multivariate analysis, older age, CQ exposure duration >5 years, and current daily dose of at least 2.5 mg/kg-IBW were the factors significantly associated with CM with OR of 5.89 (2.38, 14.57), 2.94 (1.10, 7.83), and 3.32 (1.04, 10.60), respectively while plasma CQ and DCQ showed no association with CM. With respect to plasma CQ and DCQ levels, current daily CQ dosage had the strongest positive correlation with both plasma levels as shown in the table. Conclusions Age more than 60 year-old, duration of CQ use more than 5 years and current CQ dose ≥2.5 mg/kg-IBW/day were the risk factors of CM. The plasma CQ or DCQ levels demonstrated no correlation in developing CM. The plasma levels were strongly associated with current daily dose of CQ. Funding Siriraj Research Development Fund, Faculty of Medicine Siriraj Hospital, Mahidol University. Disclosure of Interest None Declared