Epidemiology of Osteoarthritis Highlights
- Abstract #1990: Association of baseline vitamin C with incident and progressive radiographic knee OA. The MOST Study.
- Abstract #1993 Associations between uses of non-steroidal anti-inflammatory drugs, knee cartilage loss and knee cartilage defect development in older adults: the Tasmania Older Adults Cohort (TASOAC) study.
- Abstract # 701 Planus foot morphology is associated with knee pain and cartilage damage in older adults.
- Abstract #F62 Using ultrasound to predict response to intraarticular steroid injections in knee osteoarthritis.
- Abstract # 880 Do cartilage volume or thickness distinguish knees with and without mild radiographic osteoarthritis? The Framingham Study.
- Abstract # 206 Moving to maintain function in knee osteoarthritis.
- Abstract # 1373 Sleep disturbance in older women with painful radiographic hip osteoarthritis.
Abstract #1990: Association of baseline vitamin C with incident and progressive radiographic knee OA. The MOST Study.
Objective:
The authors sought to examine the association between serum levels of vitamin C to incident and progressive radiographic evidence of knee osteoarthritis.
Methods:
This analysis examined the MOST cohort, a longitudinal study of 3026 men and women aged 50-79 year with known, or at high risk to develop, knee osteoarthritis. Participants underwent serial knee x-rays, both at entry into the study and at 30 months of follow-up. Weight-bearing x-rays were obtained and then scored for individual features of osteoarthritis, including joint space narrowing and osteophytes, as well as for a global Kellgren-Lawrence grade. Control knees were randomly selected from the source population, but had no radiographic evidence of knee osteoarthritis at 30 months follow-up. Multivariate analysis was performed, with adjustment for age, gender, body mass index, nonwhite race, clinic site, assay batch, and baseline glucosamine and chondroitin supplement use.
Results:
In analyses which focused on the relationship between successively higher quartiles of serum vitamin C at baseline to incidence of radiographic knee osteoarthritis, there was no clear association present [odds ratio (OR) 0.77; 95% confidence interval (CI) 0.45 – 1.30]. In contrast, an association [n=408 knees] between quartiles of serum vitamin C with progressive knee osteoarthritis was evident. Specifically, those in the lowest quartile of baseline vitamin C displayed more than a two-fold greater risk for progressive knee osteoarthritis than those in the highest quartile level [OR 2.12; 95%CI 1.30 – 3.46].
EDITORIAL COMMENT:
Whereas dietary intake of vitamin C has been previously studied in relation to progression of knee osteoarthritis, with higher levels found to be protective against progressive disease, the relation of serum levels of vitamin D has been unknown. In this context, this study demonstrates a clear relationship between levels of serum vitamin C with progression of osteoarthritis of the knee, with osteoarthritis status assessed radiographically and serially over time. These findings have clear implication for patients with established knee osteoarthritis, and suggest a potential role for therapeutic intervention, using vitamin C supplementation, which may be successful at averting, or at least diminishing the rate of progression, of knee osteoarthritis.
Abstract #1993 Associations between uses of non-steroidal anti-inflammatory drugs, knee cartilage loss and knee cartilage defect development in older adults: the Tasmania Older Adults Cohort (TASOAC) study.
Objective:
The investigators sought to determine the association between use of non-steroidal anti-inflammatory drugs (NSAIDs), knee cartilage loss and development of knee cartilage defects in a prospective study with follow-up of ~2.9 years in a population of older adults.
Methods:
The TASOAC study consisted of 419 randomly selected subjects with a mean age of 62 years; the participants were equally composed of men and women. Knee cartilage was assessed with regard to presence of cartilage defects and volume using T1-weighted fat-suppressed MRI imaging of the right knee. Cartilage volume and the presence (or absence) of knee cartilage defects were measured at the medial and lateral tibial sites at two points in time, both at cohort entry and during follow-up, 2.9 years later. Use of conventional NSAIDs and of selective cyclooxygenase-2 (COX-2) inhibitors in the last month was recorded by questionnaire.
Results:
The users of COX-2 inhibitors (n=49) had a lesser frequency to develop knee cartilage defects in the medial tibiofemoral compartment (OR 0.39, 95% CI 0.15 - 0.99) compared to the NSAID non-users (n=346). However, there was no greater loss of cartilage volume at the medial and lateral tibial sites between these two groups. As for the users of non-selective NSAIDs (n=24), they more frequently developed knee cartilage defects in both the medial (OR: 3.07, 95% CI: 1.04 - 9.09) and lateral (OR: 2.61, 95% CI 1.01 - 6.77) tibiofemoral compartments. Once again, there was no significant difference in loss of cartilage volume between these two groups, at both the medial and lateral tibial sites. Interestingly, users of non-selective NSAIDs experienced more cartilage volume loss (β = -4.85% pa, 95% CI: -9.00, -0.70 at medial tibia; and β = -4.61%, 95% CI: -8.11, -1.12 at the lateral tibia) compared to users of selective COX-2 inhibitors.
EDITORIAL COMMENT:
Nonsteroidal anti-inflammatory drugs (NSAIDs) are routinely prescribed by physicians for the medical management of knee osteoarthritis, particularly to target the pain associated with degenerative joints. Prior investigation has attempted to elucidate whether NSAIDs have chondroprotective vs. chondroneutral vs. chondrodestructive effects on articular cartilage. Concern exists as to whether use of NSAIDs may be harmful to joints and lead to impaired chondroycte function and growth, and thus contribute to unfavorable progression of osteoarthritis, even though their use is well motivated to decrease the pain associated with osteoarthritis joints.
This study suggests that use of both non-elective and of selective NSAIDs may lead to greater frequency of cartilaginous defects. The clinical significance of such defects is, however, not clear. The results also suggest that loss of cartilage volume may be greater in non-selective NSAID users compared to selective COX-2 antagonists. It is imperative to note, however, that the number of NSAID users was relatively small. Replication in larger numbers of study participants is warranted. In addition, in this study, we are only informed as to the broad category label of selective versus non-selective NSAIDs. It would seem worthwhile to identify and study the effect, in a systematic fashion, of particular NSAID agents on cartilage volume and defects.
Abstract # 701 Planus foot morphology is associated with knee pain and cartilage damage in older adults.
Objective:
The authors sought to determine whether the presence of planus foot alignment is related to [A] cartilage damage and [B] to the presence of pain in the ipsilateral knee.
Methods:
This potential relationship was examined among older men and women in the population-based Framingham Foot and OA Studies. The nature of the participants’ standing foot morphology was measured from plantar pressure recordings. In turn these recordings are used to calculate an Arch Index, which quantitatively increases as the degree of planus foot morphology increases; a value of zero is assigned for cavus foot morphology. In addition, the presence of knee pain on most days was assessed by questionnaire. Finally, proton density-weighted sagittal and coronal MRI imaging was obtained and quantified on The WORMS scale (0-6), on which cartilage damage is measured. Using logistic regression analysis, the relative odds of ipsilateral knee pain and of cartilage damage was assessed in relation to each of four categories of the AI. Category 1 included only cavus feet, whereas categories 2-4 reflected the tertile distribution of AI among non-cavus feet. In multivariate analyses, adjustment was performed for age, sex, and BMI.
Results:
There were a total of 1884 participants among whom just over half were female. The average age was 65 years. In this group, 22% of the knees were reported as being painful on most days. Cartilage damage was present in 45% of the medial tibiofemoral compartments and among 42% of the lateral patellofemoral compartments. Feet with the greatest degree of planus morphology had a 39% increased risk of having ipsilateral knee pain and 76% times the odds of medial tibiofemoral cartilage damage, as compared with feet displaying cavus morphology. In contrast, there was no association observed between foot morphology and cartilage damage in the lateral patellofemoral compartment.
EDITORIAL COMMENT:
This is certainly an exciting study, with an entirely novel approach to study epidemiologic associations and potential pathogenetic contributions to osteoarthritis of the knee. While substantial effort has been previously directed at understanding the contribution of body weight and load forces as transmitted in a cephalad-to-caudal direction across the knee joint compartment, the present study, instead, focused on a downstream or distal joint area, namely the ipsilateral foot, and how this site might impact on cartilage damage of the ipsilateral, proximal knee.
Remarkably, and of much interest, the authors did find such an association to be present. In fact, the presence of a planus foot morphology was associated with a greater degree of ipsilateral knee pain and with medial tibiofemoral cartilage damage. A relationship with patellofemoral compartment cartilage injury was not found. It is important to recognize that the present study was cross-sectional in design, precluding any inference about the directionality of this association. A future longitudinal analysis will certainly be informative. In addition, the reader will in the future no doubt benefit from a biomechanical study which can actually measure, directly, the impact on forces across the femur and tibia that result from planus foot morphology.
Abstract #F62 Using ultrasound to predict response to intraarticular steroid injections in knee osteoarthritis.
Objective: Among a group of patients with osteoarthritis of the knee, the aim of the present study was to determine whether inflammation, as measured by ultrasonography, is predictive of clinical response to corticosteroid injection.
Methods:
The participants in this study were symptomatic patients with knee pain who satisfied clinical and/or radiographic criteria for knee osteoarthritis. Ultrasound was performed on all participants at study entry, after which the participants were randomized to receive either an injection of 40mg of triamcinolone acetonide or 1cc of a saline (placebo) solution. Clinical response was assessed using the WOMAC scoring system and by physician global assessment from baseline to weeks 4 and 12 of follow-up. Ultrasound examination was repeated at each visit. In addition, all participants were further categorized at baseline as either having (or not having) joint inflammation, as defined by the presence (vs. the absence) of sonographic evidence of synovial hypertrophy >1 mm, with or without an effusion.
Results:
The study consisted of 74 patients who were randomized, though follow-up data at the 1 month interval were only available for 66 patients. There was no significant difference in baseline WOMAC scores between the two groups. Importantly, whereas there was no change in WOMAC pain score among the control patients from baseline (mean 10.0 +/- 3.2) to 1 month of follow-up (mean 10.0 +/- 3.1), there was significant improvement observed in the active (triamcinolone) treatment injection group [WOMAC pain score decreased from 10.9 (+/- 3.1) to 8.8 (+/- 4.1); p<0.001]. In addition, there were 28 patients (42%) with sonographic evidence of inflammation disease on ultrasound. But, there was no observed difference in WOMAC pain scores between the inflamed and the non-inflamed patients in the active treatment group.
EDITORIAL COMMENT:
This study addresses an important mainstay in the medical management of osteoarthritis of the knee, namely the use of intra-articular steroid injection. The challenge to the treating physician is in the determination as to which patients might benefit from this therapeutic intervention. A priori, the authors reasoned that it might be those patients with pre-injection evidence of joint inflammation who would stand to derive benefit from this approach. However, in this study, there was no apparent difference in change in WOMAC pain scores, between the patients in the active treatment (steroid injection) group who had and those who did not have ultrasound evidence of synovitis
At the same time, it is important to knee that the study may have been underpowered to show this effect (the actual numbers are not furnished in the Abstract). Thus while the study did show that active steroid injection, as compared to placebo saline injection, does result in symptomatic improvement, it also highlights the importance of identifying which patients are best suited to be offered this treatment approach.
Abstract # 880 Do cartilage volume or thickness distinguish knees with and without mild radiographic osteoarthritis? The Framingham Study.
Objective:
The authors examined whether the quantity of cartilage, as assessed using magnetic resonance imaging, differs between those patients with mild radiographic evidence of knee osteoarthritis compared those with osteoarthritis at the knee joint.
Methods:
Participants in the Framingham Osteoarthritis Study underwent MRI evaluation, which enabled a tibiofemoral MRI-based measurement of cartilage volume to be ascertained for each member of the study. Within the knee joint, a measurement of cartilage volume, subchondral bone area and thickness and of subregional bone thickness was obtained. In addition, conventional posteroanterior fixed flexion, weight-bearing radiographs were obtained to assess for Kellgren/Lawrence (K/L) grade status (0-4). The two groups compared in the present study were those with a mild grade of osteoarthritis, i.e. Kellgren-Lawrence grade 2 (indicating the presence of definite osteophytes with mild medial joint space narrowing) and those without osteoarthritis. Potential differences in measures of cartilage structure and quantity, between those participants with and those without osteoarthritis, were then measured.
Results:
There were a total of 948 participants whose knees were evaluated; within this group, 11% had radiographic evidence of mild osteoarthritis whereas 76% were without x-ray evidence of osteoarthritis. Notably, MRI imaging did not quantitatively distinguish those with mild from those without x-ray evidence of osteoarthritis. However, among those with mild osteoarthritis, especially among men, subchondral bone thickness was lower (p<.05) than in those without osteoarthritis. Semi-quantitative scoring for focal erosions showed higher scores as well for those with mild osteoarthritis than in those with no osteoarthritis at all. Finally, in the remaining study participants, those with moderate and severe grades of knee osteoarthritis, these MRI findings were much more apparent.
EDITORIAL COMMENT:
These findings suggest that magnetic resonance imaging may assist in the early detection and differentiation of those with mild osteoarthritis from those without osteoarthritis at all. Such MR differences are more evident with regard to measures of subchondral bone thickness than with cartilage volume. These findings may be particularly beneficial in the research setting, where investigators are eager to detect the onset of osteoarthritis at its earliest stages, which may afford an improved understanding of the forces and factors that contribute to its onset, and offer an opportunity to observe which factors might predict its future course. However, these subtle MRI finding do not as yet have any well accepted or proven role in the medical care and diagnostic evaluation of patients presenting with knee pain to the doctor. The standard diagnostic radiographic measure of osteoarthritis remains the weight-bearing plain film of the knee.
Abstract # 206 Moving to maintain function in knee osteoarthritis.
Objective:
The authors sought to assess the impact of physical activity on functional outcomes among persons with knee osteoarthritis.
Methods:
This study focused on a publicly available arthritis database, namely the Osteoarthritis Initiative. Such data were available on 1265 adults with baseline x-ray evidence of knee osteoarthritis who were aged 45-79 years. The participants underwent assessment both at baseline and at 1-year of follow-up. The functional outcomes of interest included the Short Form 36 physical component score (PCS), chair stand rate, and 20 meter walk rate. Specifically, physical activity scores were categorized along several domains, including (1) walking, (2) purposeful sports or recreation, and (3) other activity components. As such, the authors focused on the predictive value of each activity group at baseline on the likelihood of good function level at 1-year of follow-up.
Results:
Importantly, more than 1 in 3 participants with radiographic evidence of knee osteoarthritis improved or maintained high function at 1-year of follow-up. Notably, walking, at study entry, contributed 9% and sports/recreation contributed only 6% to the total Physical Activity Scores for Elderly. Importantly, participation in sports/recreational activities was related to a greater likelihood of good function during 1 year of follow-up, than in those participants not engaging in such physical activity at study entry.
EDITORIAL COMMENT:
These data add further support as well as evidence to the notion that patients with symptomatic and radiographic evidence of knee osteoarthritis ought not to be encouraged to adopt a sedentary lifestyle. In fact, being active leads to an improvement, rather than a diminution, in physical function in those with symptomatic and radiographic evidence of osteoarthritis at the knee joint. These findings have important implication for the messages doctors ought important to their patients with radiographic evidence of symptomatic knee osteoarthritis, when patients present with queries about how to optimize their functional capacity.
Abstract # 1373 Sleep disturbance in older women with painful radiographic hip osteoarthritis.
Background:
The authors examined the potential association of impaired quality sleep with radiographic evidence of osteoarthritis, at either the knee or hip joint.
Methods:
This potential association was evaluated using the Study of Osteoporotic Fractures. In the first component of the present study, the authors analyzed 90 women with painful radiographic evidence of hip osteoarthritis as well as 3429 women without hip pain or osteoarthritis. Each participant furnished information regarding self reported difficulty maintaining sleep. Osteoarthritis of the hip was assessed from pelvic radiographs. In addition, in the second sub-study, 362 women with self-reported moderate to extreme hip pain when walking as well as 2409 women with less or no pain were assessed for sleep quality and for wakefulness after sleep onset, as assessed by wrist actigraphy.
Results:
The study participants in the first component analysis were aged ≥ 81 years. As suspected, women with, as compared to those without, painful radiographic evidence of hip osteoarthritis were two-times more likely to experience difficulty with sleep maintenance (age-adjusted OR=2.1; 95%CI: 1.3 - 3.3). In the second component study, in multivariate models adjusted for age, race, BMI, RA and non hip pain, the authors observed a 3% (95% CI: -4.6, -1.1) reduction in mean sleep efficiency and 8.4 (95% CI 3.0, 13.8) minute increase in total time of wake after onset of sleep among those women with moderate-to-extreme hip pain while walking. Importantly, the observed associations persisted, for both sleep measures remained significant after adjustment for health status, medication use and depression (p =0.02).
Editorial Comment:
The authors convey an important and infrequently noted finding, namely that older adults with radiographic evidence of hip osteoarthritis experience a greater, on the order of a two-fold increase in risk, of hip osteoarthritis than their peer counterparts without hip pain. These findings are important in drawing attention to the apparent association of impaired quality sleep with radiographic evidence of symptomatic hip osteoarthritis. One need note that there was an apparent two-year interval in time between the time that the pelvis radiographs were obtained and when the study participants were queried for the presence of hip pain. It would be more optimal to obtain the radiographic and symptomatic status of the participants at the same point in time, or as close to it as is feasible.
