Role of Alphacalcidol to Reduce Pain and Serum Cartilage Oligomeric Matrix Protein in Elderly with Knee Osteoarthritis

Objective : To determine the effect of Alphacalcidol supplementation on pain based on WOMAC pain index and joint cartilage condition based on cartilage oligomeric matrix protein (COMP) serum markers in knee osteoarthritis (OA) elderly patients. Materials andMethods : This is a randomized, double-blind, and placebo-controlled trial. Symptomatic knee OA elderly subjects visited our clinic in Jakarta July-December 2017 were recruited. History taking, physical examinations, and knee radiology were performed. Serum Vitamin D (25(OH)D) and COMP were analysed using chemiluminescent immunoassay and enzyme-linked immunosorbent assay, respectively. Body mass index, physical activity, sun exposure frequency, and dietary vitamin D were assessed. Subjects were randomly allocated to either intervention group to be given 1 µg Alphacalcidol or control group to be given placebo for 12 weeks. Comparation analysis of WOMAC pain index and serum COMP concentration between both groups was performed. Results : There were 146 subjects participated this study. Increases of 25(OH)D were found in intervention group (2.63 ±11.24 nmol/L, p=0.05) and control groups (1.09±9.42 nmol/L, p=0.28). Alphacalcidol significantly reduced pain based on WOMAC indicator with mean reduction differences of intervention group compared to control group was 2.174±1.060 (p=0.00). COMP serum level was reduced with mean reduction differences of intervention group compared to control group was 38.15±87.553 ng/ ml (p=0.39). Alphacalcidol and gender were the determinants of WOMAC pain index reduction (p=0.00 and p=0.06, respectively) while Alphacalcidol was the only serum COMP level change determinant (p=0.39). Conclusion s :Alphacalcidol administration reduced pain based on WOMAC indicator and COMP serum in knee OA elderly subjects.


Introduction
Indonesian elderly population has been increasing in the last 30 years. 1,2 Life expectancy of Indonesian population has also been continuously increasing to 69.3 and 73.19 years old for male and female, respectively, in 2017. In 2095-2100, it is estimated to reach 82.47 and 84.97 years old for male and female, respectively. 3 Increasing elderly population potentially increases the number of degenerative diseases.
Osteoarthritis (OA) is the leading cause of mobility impairment. It is the fourth of the most common Years Lost to Disability (YLD) globally, which contributes to 3% of global YLD. 4 Knee OA accounted for 36,81% cases of internal medicine clinic at Manado Central Province Hospital during March 1994 -November 1995. 5 Other studies at Sanglah Denpasar Hospital in [2001][2002] showed that OA contributed to 37% of all rheumatic cases. Around 97% subjects in the study had knee OA. 5 However, national prevalence of OA in Indonesia is still not available.
Vitamin D plays important role to enhance calcium absorption in intestine and promotes mineralization of bones. 6 Several studies found that Vitamin D levels were low in patients with knee OA. Bassiouni et al found statistically lower mean serum level of vitamin D in OA patient group than in controxls (8.64 ± 6.42 vs 14.84 ± 0.87 pg/mL, p =0.0295). 7 Other study by Veronese et al showed that lower 25OHD quartiles were associated with the presence of knee OA (OR = 1.17, 95% CI 1.05-1.29). 8 Study by Jansen and Haddad showed high prevalence (around 24% of patients) low vitamin D levels in OA patients. 9 Vitamin D deficiency were found in 50% of women age 45-55 years old and 35% of women age 60-90 years old in Jakarta and Bekasi. 10,11 Alphacalcidol is a vitamin D supplementation recommended to be used in patients with renal insufficiency, which is prevalent in elderly. Alfacalcidol does not need to undergo renal 1α-hydroxylation which impaired in renal insufficiency, although it can still cause hypercalcemia. 12,13 Current detection and diagnosis of knee OA relies on plain radiological examination, which is not really objective because it depends on radiological expertise and experience. Moreover, plain radiological diagnosis has limited ability to detect joint damage at an early stage. Therefore, knee OA is often diagnosed at advanced stage. For this reason, selecting accurate joint cartilage biomarkers is important to detect and predict the severity of OA objectively and at earlier stage. Cartilage Oligomeric Matrix Protein (COMP) is one of degradation products released into synovium fluid and bloodstream from cartilage matrix turnover process. In a study conducted by Andriyasa and Putra, more severe knee OA tend to have higher serum COMP concentration, although there was no statistically significant correlation (r=0.127, p=0.31). 5 The CHECK (Cohort of Hip and Knee) study conducted by Van Spil et al looked at the relationship between systemic biochemical markers of joint metabolism and inflammation on radiographic parameters and pain. It was found that COMP serum as a marker of cartilage degradation was positively related to osteophyte area (stand beta -0.092, p=0.005) but not to joint space width. 14 Another study by Bartels et al observed changes in biomarkers of bone and cartilage in knee OA patients after intensive weight loss interventions for 16 weeks. The result showed mean weight loss of 13.4 kg (95% CI: 12.5 -14.4) and serum COMP reduction of about 1.1 (95% CI: 1.5 -0.8) U/L which correlated to weight loss (r = -0.17, p = 0.028). 15 Therefore, COMP can be used as a marker of diagnosis and cartilage injury severity of knee OA. 8 Currently, there has been no study about the association of Vitamin D supplementation with pain level and cartilage matrix turnover process in Indonesia. Therefore, this study aims to determine the effect of Vitamin D supplementation (Alphacalcidol) on pain and joint cartilage condition in elderly patients with knee OA.

Study design
This is randomized, double-blind, and placebocontrolled clinical trial. The effect of Vitamin D supplementation (Alphacalcidol) on pain based on WOMAC index and on serum COMP level changes would be studied in elderly patients with knee OA. This study was started on January 2017, starting with planning and permission obtaining. The diagnosis of knee OA was made by clinical examination and x-ray. In this study, only WOMAC pain index was considered because pain is the only subjective sensations studied. In Indonesia, it is common to see elderly people came to outpatient clinic with pain as the main symptom. COMP is considered as the objective conditions of joint cartilage condition while WOMAC index is the subjective variable for detecting knee OA. Other variables that potentially affected this study were analyzed, including age, gender, body mass index (BMI), physical activity, knee radiology grade, 25(OH)D serum level, sun exposure frequency, and Vitamin D consumption. This study used serum level of 25(OH)D as the vitamin D adequacy indicator since individual vitamin D status is better reflected with 25(OH)D than its active form, 1,25(OH) 2 D (calcitriol). The level of calcitriol is tightly regulated and maintained within its reference values even though individual actually had vitamin D deficiency. [16][17][18] The half-life of 25(OH)D is also longer than calcitriol, which is 2-3 weeks and 4-6 hours, respectively, making 25(OH)D a better marker of vitamin D adequacy. [17][18][19] This study is monitored by Contract Research Organization (CRO), The Equilab International, to ensure validity and carry out Good Clinical Practice (GCP) and Good Laboratory Practice (GLP).

Participants
Subjects with symptomatic knee OA pain who came to our clinic (Primary Health Care Clinic of UIN Jakarta) during July -December 2017 were recruited by consecutive sampling. Subjects examinations included history taking, physical examinations, knee radiology, and blood test for serum 25(OH)D, calcium, and COMP. Subjects were later divided into two groups, intervention and placebo groups.

Procedures and Variables Definitions
Serum 25(OH)D was measured using chemiluminescent immunoassay. Subjects were asked to not take any meal for 12 hours prior to 25(OH)D measurement. Serum COMP was measured using BioVendor RD194080200 Human Cartilage Oligomeric Matrix Protein Enzyme-Linked Immunosorbent Assay (ELISA). As many as 10 mL blood was obtained to measure serum COMP after at least 3 hours of sleep and 60 minutes of rest. Physical activity level was categorized using Indonesian Basic Health Research Questionnaire 2007. 20 Mild physical activity was defined as sedentary activity, including sitting, lying down, reading, playing games, or watching TV which was done for >6 hours a day. Moderate physical activity was defined as sweeping the yard, mopping, washing clothes, drawing water, planting, cleaning the bathroom or pool, playing tennis, badminton, aerobic gymnastics, swimming, basketball, soccer, or similar activity for 4-7 days a week and >60 minutes per day. Intense physical activity was defined as lifting wood, rice, sandstone, hoeing, cutting trees, fast-cycling, weightlifting, sprinting, marathons, pedicab rides, mountain climbing, or other similar activities for 4-7 days a week and >60 minutes per day. Osteoarthritis knee radiology was based on Kellgren-Lawrence knee OA grading. 21 Sun exposure was defined as ultraviolet (UV) B exposure to at least face and palms between 7.00 a.m. and 11.00 a.m. local time for 30 minutes. UVB exposure was measured using UV meter as in another study by Setiati et al. 11 Mean vitamin D consumption was assessed using Food Frequency Questionnaire (FFQ) every 4 weeks for 12 weeks.
All blood test procedures were performed by certified personnel and laboratory.
Subjects were educated to have UVB exposure in face and arms area for 30 minutes at 9 AM 3 days a week without using sunblock or umbrella. Subjects were routinely contacted using telephone call by our clinic personnel every 7 days to ensure that the vitamin D supplement or placebo was taken daily and to remind the next follow up schedule. Subjects in both groups were monitored for patient adherence and side effects every two weeks in our clinic. WOMAC pain score, physical activity, sun exposure, and adverse events were monitored and reassessed. Semi quantitative Food Frequency Questionnaire (FFQ) was assessed every 4 weeks. After 12 weeks, subjects will be reexamined for changes in WOMAC pain index, knee radiography, and blood examination for 25(OH)D and COMP serum level.

Inclusion and exclusion criteria
To be eligible for becoming participant, individuals had to be around 60 to 79 years old, diagnosed as knee OA from clinical and radiological criteria, had minimum pain score of 5 (WOMAC pain index), BMI ≤27 kg/m 2 , and were not suffering from systemic inflammation condition or disease. Using minimum WOMAC pain index score 5 inclusion criteria was done on researcher discretion because there was no standardized limit on patient inclusion based on the WOMAC pain index. Consumption of paracetamol as rescue medication was allowed in the presence of increased pain. If the pain was not reduced after oral paracetamol administration, the subject was excluded.
The exclusion criteria in this study were grade 4Kellgren-Lawrence, WOMAC pain index >15, serum level of 25(OH)D >125 nmol/ liter, serum calcium level >10.5 mg/dl, other kind of arthritis or inflammatory diseases, conditions that disrupt oral absorption, knee trauma history, conditions that will interact with Vitamin D, consuming supplement or steroid, undergo intraarticular therapy, and dementia. Subjects with grade 4 Kellgren-Lawrence were excluded because there was structural damage in the form of bone-on-bone contact with cartilage loss which required surgical therapy, not just limited to conservative therapy. 22,23 Randomization and subject allocation Subjects fulfilled inclusion and exclusion criteria were randomly allocated to intervention or placebo group using computer-generated block permutation randomization comprising four participants per block. This procedure was performed by CRO. Group allocation was kept by CRO to ensure double-blind principle was performed. Individual subject's number of orders was informed to pharmacy depo. Individuals in intervention group were given oral Vitamin D supplements (Alphacalcidol1 µg) once daily while individuals in control group were given placebo once daily. Beside this intervention, there was no difference of any other treatment. Both groups continued the intervention for 12 weeks.

Statistical analysis
The number of subjects required was calculated using hypothesis testing for two population means by Lemeshow et al. 24 The calculation was performed twice, using both WOMAC pain score and serum COMP concentration, as separate outcome. In this study, significant mean difference of WOMAC pain score was assumed to be 2 obtained from McAlindon et al study while significant mean difference of serum COMP concentration was assumed to be 2 obtained from Forsbladd'Elia et al study. 25,26 Power assumption and confidence interval used were 80% and 95%, respectively. The number of subjects required to determine the effect of alphacalcidol supplementation on pain based on WOMAC pain scale was 68 for each arm. The number of subjects required to determine the effect of alphacalcidol supplementation on COMP serum marker was 38 for each arm. Dropout rate was estimated at 10%. Therefore, the number of subjects required for this study was 150 subjects. Data collected were analysed using SPSS for Windows 21 st edition. All numerical data were presented in mean ± deviation standard. Significant differences of subjects' characteristics were determined using Chi-Square test. Comparation analysis of WOMAC pain index and serum COMP concentration between intervention group and placebo group was performed using either independent T-test or Mann-Whitney test. In the end, variables affecting WOMAC pain index and serum COMP concentration were determined using multiple linear regression.

Results
One hundred and fifty subjects participated at the start of the trial and 146 (97.3%) subjects finished the trial. Only participants completing the study were included in data and be analysed further. Subjects were recruited and followed up in July -December 2017. The subject diagram flow for this trial can be seen in figure 1 below,

Fig 1. Subject Diagram Flow
In this study, majority of the subjects were young elderly, women, had normal BMI, mild physical activity, and had osteoarthritis grade KL 1-2. Most of the subjects were normocalcemia and exposed to sunlight ≥3times weekly. There is quite even distribution in Vitamin D consumption. The baseline characteristics are shown in table 1 below, Vitamin D supplementation (Alphacalcidol) and gender were the determinant variables of WOMAC pain index reduction (Table 4). Interaction between gender and Vitamin D supplementation variable was not found (p>0.05). Vitamin D supplementation variable was the only determinant variables of serum COMP level changes but not statistically significant (p=0.39) ( Table 5). During this study recruitment and follow up period, no subjects experienced any adverse event nor side effects.     Bangladesh (43.8%), with also wider spectrum of age group. 28 There is a tendency that serum 25(OH)D level is lower as subject gets older.
In this study, alphacalcidol was used because it does not need to undergo 1a-hydroxylation process, which is impaired in patients with renal insufficiency. 12,13 In addition, the effects of hypercalcemia are not higher than other vitamin D supplementations. 13 41 This study also complements metanalysis by Wu et al which observe the effect of Vitamin D supplementation on changes in the degree of chronic pain. 42 The study results in a significant pain reduction in intervention group compared to placebo.
Vitamin D has been proven to stimulate proteoglycan synthesis in joint cartilage in vitro. 43 In this study, serum COMP levels reduction was observed in intervention group. Meanwhile in control group, serum COMP level was increased. No statistically significant difference in mean differences between serum COMP levels of pre and post-intervention interventions between both groups. It can be caused by the amount of dose and duration of Vitamin D supplementation that is not large and long enough which cannot produce a statistically significant difference yet.
No statistically significant serum COMP levels change was observed across several variables as shown in Table 3. It supports several studies published previously. [44][45][46][47] Despite that, studies related to COMP and the variables until this time show contradictive results. There are no studies relating Vitamin D level with serum COMP level change in knee OA up until this study was done. However, the results obtained in final model showed that supplementation of Vitamin D (Alphacalcidol) for 12 weeks contributed to the improvement of the knee joint cartilage as indicated by a decrease in serum COMP level after intervention.
This is the first study of knee OA in Indonesia which combines symptoms modifier indicator (WOMAC pain indicator) with disease modifier indicator (joint cartilage damage). Therefore, the protocol of this study was designed by authors. There are several limitations of this study. The first limitation is difficulty in objectively assessing the degree of pain becomes a potential information bias. Information bias may also occur in the process of filling semiquantitative FFQ. However, the authors had put efforts in maintaining the originality and acceptability of the result of this process by training the interviewers and informing subjects about this process. In addition, measurements of COMP concentrations carried out through blood serum collection may be influenced by the presence of non-clinical OA or RA disease processes in respondents.

Conclusion
Vitamin D supplementation (Alphacalcidol) for 12 weeks affects the reduction of pain level based on WOMAC indicators and the reduction of serum COMP levels in elderly with knee OA although the reduction of serum COMP was not statistically significant. Vitamin D supplementation can benefit the elderly with early knee OA and Vitamin D insufficiency to reduce the pain of the knee joint.

Ethics Approval
The procedures followed were in accordance with the ethical standards of the responsible committee on