Osteoarthritis and Cartilage 27 (2019) 359e364

Review

Osteoarthritis year in review 2018: clinical

L.A. Mandl y z *
y Hospital for Special Surgery, USA
z Weill Cornell Medicine, USA

a r t i c l e i n f o s u m m a r y

Article history: Osteoarthritis (OA) is the most common joint disease in the world, with an age-associated increase in
Received 24 July 2018 both incidence and prevalence. Clinical and epidemiologic research is crucial to better understand risk
Accepted 8 November 2018 factors for disease, find the best treatments for symptoms, and identify therapies to slow down or even
prevent disease progression. This paper is based on a systematic review of the osteoarthritis literature
Keywords: published in English between 2017/05/01 and 2018/04/25, with a focus on papers which have the po-
Osteoarthritis
tential to improve patient care, or which suggest novel areas for future research.
Epidemiology
© 2018 Osteoarthritis Research Society International. Published by Elsevier Ltd. All rights reserved.
Clinical

Introduction Arthritis and Rheumatology, Arthritis Care and Research, Annals of

the Rheumatic Disease and The Journal of Rheumatology and
Osteoarthritis (OA) is a leading cause of disability in the United contained the search term “osteoarthritis”. This resulted in 171
States with over 22.7 million people reporting arthritis-attributable articles, with some overlap. Titles were reviewed, and papers
activity limitations1,2. Only depression and alcohol misuse result in excluded if their primary focus was non-clinical, a case-series, a
more years lost to disability3. Already high, the incidence and description of a study protocol, or were best aligned with one of the
prevalence of OA is predicted to skyrocket over the coming decades other Year-in-Review content areas. Reference lists of select articles
due to the aging population, rising obesity rates and high rates of were hand searched for additional potential articles of interest.
traumatic knee injuries4e7. This is a public health crisis, and there is It is of course impossible in this brief review to discuss every
a pressing need for rigorous high-quality OA clinical research to important osteoarthritis manuscript published in the last year. The
ensure patients receive safe and effective treatments. This paper is choice of which articles to highlight was based on the journal impact
a subjective overview of some of the most notable osteoarthritis factor, the potential impact of the study on patient care, the impact
clinical research studies published in the last year. on the study on furthering novel areas of research, and opinions
solicited from experts in the field of osteoarthritis clinical research.
Methods
Incidence, prevalence and progression of OA
A PubMed search was performed for articles published between
2017/05/01 and 2018/04/25. Search terms were (osteoarthritis AND The associations between older age, obesity and increased rates
treatment) OR (osteoarthritis AND therapy) OR (osteoarthritis AND of knee OA are well understood. However a study performed by
epidemiology), with results limited to English language studies Wallace et al. suggests that these major risk factors are insufficient
evaluating human subjects. Including articles listed as [Epub ahead to explain the exponential increase in the prevalence of knee OA8.
of print], this resulted in 1673 references. In addition, a comple- This group utilized skeletal samples of adults over age 50 who lived
mentary PubMed search was performed for the same date range for in urban areas in the United States. They compared skeletons from
articles which were published in the New England Journal of people who died between 1905 and 1940, (early industrial;
Medicine, Annals of Internal Medicine, Osteoarthritis and Cartilage, N ¼ 1,581) and those who died between 1976 and 2015, (post-in-
dustrial; N ¼ 819). They also included a comparator group of pre-
historic skeletons from archeologic sites in North America,
* Address correspondence and reprint requests to: L.A. Mandl, Rheumatology,
Hospital for Special Surgery, 535 East 70th Street, New York, NY 10021, USA. Tel: 1-
(N ¼ 176). They found that prehistoric and early industrial skeletons
212-774-2044; Fax: 1-646-714-6304. did exhibit evidence of knee osteoarthritis. However, since the mid-
E-mail address: [email protected]. twentieth century, the prevalence of knee osteoarthritis has

https://doi.org/10.1016/j.joca.2018.11.001
1063-4584/© 2018 Osteoarthritis Research Society International. Published by Elsevier Ltd. All rights reserved.
360 L.A. Mandl / Osteoarthritis and Cartilage 27 (2019) 359e364

doubled compared to early industrial rates, even after controlling Therapies for OA
for age and body mass index, (BMI). This finding suggests in-
teractions with the modern environment may play a pathogenic Probably the highest impact clinical OA paper this past year was
role in the development of osteoarthritis. These authors hypothe- a paper by McAlindon et al. evaluating intra-articular corticosteroids
size decreased physical activity could be one important factor, as it for the treatment of knee osteoarthritis14. This paper was ranked as
can a result in chronically underloaded joints with lower proteo- the fifth top article published in JAMA in 2017. In this blinded ran-
glycan content, and weaker muscles which are unable protect and domized controlled trial subjects with knee OA were administered
stabilized joints. One potential bias is that It BMI was measured at intra-articular triamcinolone or saline placebo every 3 months for
death. If there was systematic weight loss prior to death, using end- 2 years. Given that synovitis in known to be associated with wors-
of-life BMI would not accurately control for lifetime BMI, and thus ening of structural damage in knee OA, it was hypothesized that
potentially underestimate the contribution of obesity to the local treatment of synovitis may retard disease progression.
development of osteoarthritis in the modern age. Regardless, these Although knee pain and function improved in both groups, there
data underscore that knee osteoarthritis may be more modifiable was no difference in pain between the groups at the end of the 2-
than previously assumed and suggests that environmental or year study period. However, cartilage thickness as measured by
ecologic risk factors are novel areas for ongoing research. magnetic resonance imaging (MRI) was slightly decreased in pa-
Investigators used subjects from the Osteoarthritis Initiative tients receiving the intra-articular triamcinolone, with a between-
(OAI) who did not have evidence of radiographic knee OA but were group difference of 0.11 mm (95% CI, 0.20 to 0.03). These re-
at high risk for developing knee OA, to evaluate whether evidence sults suggest that rather than retard cartilage destruction, intra-
of the tissue lesions were predictive of developing incident radio- articular triamcinolone may accelerate cartilage destruction, and
graphic osteoarthritis9. Sharma et al. found that abnormalities such that the anti-inflammatory effects of steroids, (at least in the short-
as bone marrow lesions, meniscal extrusion, meniscal tears and term), are not operating as a disease modifying agent. This differ-
cartilage damage increased the probability of developing incident ence is unlikely to be clinically meaningful, however, as this is
knee radiographic knee OA over the following 7 years, over and similar to the degree of cartilage loss seen in patients who do not
above standard known risk factors. These data suggest there may show any progression of clinical or radiographic OA, (mean change
be a “window of opportunity” to intervene in certain high-risk of 0.12 ± 0.28)43. These findings contrast with those from a pre-
patients before the development of clinical or standard radio- vious similarly designed study which did not show any negative
graphic evidence of knee OA. structural impact of steroid injections15. However, this earlier study
Most radiographic studies of knee OA evaluate the tibiofemoral used plain radiographs to evaluate structural outcomes, and it is
joint, although the patellofemoral compartment can also be likely radiographs are not sensitive enough to pick up very small
affected. A study by Lankhorst et al. utilizing The Cohort Hip and changes in cartilage volume. Although these data suggest that
Knee Study found that subjects with mild symptoms of early knee regularly scheduled use of intra-articular steroids may be detri-
OA are most likely to have involvement of the patellofemoral joint mental to cartilage health, it is important for clinicians to realize
first, and then progress to combined patellofemoral and tibiofe- that these data do not suggest that periodic use of intra-articular
moral osteoarthritis10. These data suggest clinicians should corticosteroids for flares of OA pain is either contraindicated or
routinely evaluate the patellofemoral joint in patients complaining ineffective.
of knee pain, especially in the absence of tibiofemoral joint space Another study evaluated a new extended-release formulation of
narrowing on plain radiographs. Identifying this anatomic variant triamcinolone acetonide in which the steroid is delivered inside
of OA is clinically important, as there are manual and targeted microspheres which are designed to maintain prolonged concen-
physical therapy approaches specifically designed for patients with tration of the steroid within the joint16. This double-blind phase
patellofemoral OA.11 11b trial evaluated the effect of one injection of this steroid prep-
Two papers by Davis et al. suggest that certain patients are at aration on mean average daily pain over 12 weeks in patients with
risk for accelerated osteoarthritis, and that these patients are moderate to severe knee OA. Although the extended-release steroid
more likely to have knees replacements. Using the data from the formulation did not lead to better pain control at 3 months, the
OAI, these investigators identified patients who progressed from results suggested it may provide patients with more rapid onset of
having no radiographic evidence osteoarthritis to having Kellgren pain relief. This medication holds promise for patients for whom
and Lawrence grade 3 or 4 osteoarthritis within 48 months-quite systemic absorption of corticosteroids could be particularly detri-
a dramatic change12. Subjects who developed “accelerated” knee mental. Further studies are needed to evaluate whether this
OA had a specific constellation of symptoms noted at the index medication can maximize analgesia and minimize steroid side ef-
visit 1-year prior compared with those who did not have rapidly fects in high risk population such as diabetics or the elderly.
accelerating osteoarthritis. These subjects had more trouble lying There were several papers evaluating therapies which were
down, more pain when they straightened their knee, and more borrowed from other musculoskeletal conditions.
pain with walking. These subjects also reported more frequent Three studies evaluated therapies routinely used in inflamma-
pain, more frequent knee swelling, and were more likely to tory arthritis as potential therapies for erosive hand osteoarthritis.
restrict their activities due to pain. Having accelerated OA was not Two randomized double-blind, placebo-controlled trials evaluated
benign, as these patients were approximately 25 times more hydroxychloroquine17,18. Both trials were negative, providing no
likely to have a knee replacement within 9 years compared to evidence that hydroxychloroquine is effective in improving pain in
patients with radiographic knee OA which was not rapidly pro- this patient population. These studies did not use MRI to evaluate
gressive13. Whether screening for patients in clinical practice who synovitis, which would have allowed them to stratify patients by
present with this constellation of symptoms would identify pa- degree of inflammation. Therefore, these studies did not rule out
tients at high risk of OA progression is unknown. In addition, the possibility that patients with erosive osteoarthritis and high
further research would be needed to determine if knowing they levels of synovitis may preferentially benefit from hydroxy-
were at high risk for rapidly progressive OA would motivate pa- chloroquine. This patient subset could be the focus of a future
tients to aggressive pursue effective interventions to retard OA randomized trial. A third randomized double-blind placebo-
progression such as physical therapy to strengthen articular controlled crossover trial evaluated whether adalimumab was
musculature, or weight loss. effective in treating the pain associated with erosive hand
 L.A. Mandl / Osteoarthritis and Cartilage 27 (2019) 359e364 361

osteoarthritis. Adalimumab or identical placebo was given subcu- Internet based therapies
taneously every 2 weeks for a total of 12 weeks19. This was followed
by an 8-week washout period, at which time patients crossed over There are significant barriers to accessing osteoarthritis care
to the other treatment arm for 12 weeks. All subjects had to have an based on geography, cost, mobility limitation or a dearth of quali-
index joint which showed signs of active synovitis on MRI , so as to fied providers. A number of studies this year investigated whether
enrich the sample with patients who had active inflammation and internet-based interventions, which could be used to overcome
thus might be more likely to respond to adalimumab. However, such barriers, are effective in patients with OA.
results showed no improvement in pain, synovitis or bone marrow Allen et al. published a pragmatic randomized controlled trial
lesions suggesting that regardeless of the presence of active syno- that enrolled physically inactive patients with symptomatic
vitis this tumor necrosis factor inhibitor is not an effective treat- radiographic knee OA, and compared the effects of a standardized
ment for erosive hand OA. in-person course of physical therapy sessions with an internet-
Three studies evaluated whether bisphosphonates, standard based exercise program which could be accessed from home,
treatments for osteoporosis, might be beneficial to patients with and used patients placed on a wait list as controls26. The primary
OA. Since subchondral bone remodeling and bone turnover un- outcome was total Western Ontario and McMaster Universities
derlies both the pathogenesis of and pain associated with osteo- Osteoarthritis Index (WOMAC) score at 4 months. Neither the
arthritis, it's possible that bisphosphonates could both prevent internet-based exercise program nor the in-person physical
osteoarthritis progression and potentially treat its associated pain. therapy was superior to the wait list in improving knee OA
Two studies used large cohorts to evaluate the effect of symptoms. These results were a little surprising given the known
bisphosphonates on knee OA, using total knee replacement, (TKR) benefit of both exercise and physical therapy for the pain of knee
as a proxy for severe symptomatic knee OA. Neogi et al. evaluated OA. The null result may have been because there was only
older women in the UK who started bisphosphonates after being moderate uptake of the internet intervention. Only 80% of the
diagnosed with knee OA. After controlling for potential con- internet group logged on to the study website, with mean num-
founders, patients starting bisphosphates were 24% less likely to ber of days logged on being only 20.7 over the entire 4-month
have a TKR over a 3-year period, compared to similar patients who study period.
did not start a bisphosphonate20. Fu et al. utilized a large national Another randomized controlled trial by O'Moore et al. evaluated
insurance database in Taiwan to ask a similar question21. These an internet administered program of cognitive behavioral therapy
investigators evaluated rates of TKR in osteoporotic patients with (CBT) for depression in subjects with knee OA27. In this unblinded
knee OA, comparing rates between those who did and did not start study of subjects with major depressive disorder and knee OA,
a bisphosphonate. Over 2 years, patients who were adherence with subjects were randomized to six on-line sessions of a validated CBT
taking bisphosphonate had a 44% reduction in TKR compared program vs usual care. Primary outcomes were self-reported
to bisphosphonate noneusers. In addition, patients who used depression severity and the general psychologic distress at
bisphosphonates had significantly less pain medication. In contrast, 3 months. Secondary outcomes included pain, function and arthritis
a systematic review and meta-analysis of randomized controlled self -efficacy. At 3 months, 84% of subjects receiving CBT no longer
trials evaluated over 3000 subjects who received oral bisphosph- met diagnostic criteria for depression vs 50% of usual care. In
onates and found that bisphosphonates neither improved pain nor addition, WOMAC subscales and arthritis self-efficacy also
prevented radiographic progression of knee OA22. However, they improved. This study was encouraging, as it suggests CBT can be
could not rule out a potential benefit in patients with bone marrow effectively administered on-line to depressed patients with knee
lesions, as such patients have higher rates of subchondral bone OA, and that both mental and physical benefits can be maintained
turnover which may put them at higher risk for OA. Whether the for at least 3 months. The fact that CBT was effective in this popu-
contradictory findings between articles are due to unmeasured lation is heartening, as depression can be a deterrent to positive
confounders, confounding by indication or differential patient se- behaviors such as increasing physical activity and weight loss,
lection is not clear; however, at a minimum it does not appear which are known to improve knee OA pain. There was however no
bisphosphonates are harmful to patients with OA-at least in the attention control in this unblinded study, so a significant placebo
short term. Perhaps larger cohort studies with longer follow-up or effect cannot be ruled out.
future randomized controlled trials will resolve this issue. Is intriguing to hypothesize why an internet-based CBT program
An intriguing new therapy for the treatment of painful knee was effective but an internet-based exercise program was not. Allen
osteoarthritis, borrowed from interventional radiology, is genicu- at al hypothesize their aggressive case finding strategies may have
late artery embolization, (GEA). GAE is used to treat recurrent resulted in subjects enrolling in their study who were not moti-
hemarthrosis after TKR. Excessive post-operative bleeding is vated to comply with the exercise intervention26. In their study
believed to be due to synovial neoangiogenesis23. Embolization of subjects only logged on to the study website for approximately 17%
the geniculate arteries supplying blood to the areas of pathologic of the study period. By contrast, 84% of the CBT subjects completed
hypervascularity “devascularize” the synovium and thus stop the all online lessons and 40% of the CBT patients were already on
intra-articular bleeding. Since there is clear evidence that synovial antidepressant when they entered the trial. This suggests that a
inflammation is associated with pain in knee OA24, Okuno et al. large portion of the CBT subjects were already in the “action” phase
hypothesized that targeted infarction of synovium may therefore of the “readiness for change model.”28 Perhaps subjects in studies
decrease pain in subjects with knee OA. An open label cohort study of internet-based interventions need to be specifically screened for
of 72 subjects with painful knee OA underwent GEA. Subjects had a motivation to engage with an intervention when it is being
significant and clinically meaningful improvement in pain relief 24 administered remotely. This will be crucial to future studies of
months after the procedure, with no significant adverse events25. osteoarthritis interventions, as lack of trained providers and
Whether this procedure is safe and effective in more heterogeneous geographic remove between patients and providers limits real life
groups of patients, and whether it could decrease the rate of knee access to many potentially helpful interventions. Understanding
OA progression, remains to be proven in randomized controlled psychosocial attributes which influence engagement will be an
trials. important aspect of future population-based internet trials.
362 L.A. Mandl / Osteoarthritis and Cartilage 27 (2019) 359e364

Regenerative medicine despite technical success32,37. Whether limiting opioids pre-

operatively can affect long term pain relief remains to be seen.
The use of regenerative medicine therapies to treat osteoar-
thritis is an area of tremendous interest, as is evidenced by the Diet and OA
surge of publications evaluating therapies such stem cell treat-
ments and platelet rich plasma (PRP) injections. Systematic reviews Two related papers examined the association between fiber
are therefore helpful to make sense of the multiplicity of publica- intake and knee OA. The first evaluated knee pain trajectories in
tions. Amongst many recent systematic reviews of stem cell ther- subjects enrolled in the OAI39. These patients all had or were at risk
apy for OA, only one, by Pas et al., was identified as having low risk of developing knee OA. Over 8 years, patients who consumed more
of bias29,30. This systematic review evaluated randomized and non- dietary total or cereal grain fibers were less likely to have moderate
randomized controlled trials (RCTs) of different stem cell therapy or severe pain. These findings were even more pronounced among
for knee OA The authors identified five RCTs, each of which reported the patients with radiographic knee OA. The second paper by the
positive results of stem cell therapies. However, each of these same group evaluated the association between fiber intake and
studies was at high risk of bias for multiple reasons, including incident radiographic knee OA among subjects in the OAI and the
inadequate blinding, high risk of selection bias, and no intention to Framingham offspring OA study40. In both of these prospective
treat analyses, casting serious doubts their efficacy claims. A cohorts, even after controlling for confounders, there was a statis-
narrative overview by Bennell et al. identified RCTs evaluating the tically significant dose-dependent inverse relationship between
effects of PRP, 15 in knee OA and 3 in hip OA. Most studies found a total dietary fiber and developing symptomatic knee OA. Although
benefit of PRP. However there were multiple issues with quality in mechanisms are speculative, these findings could be due anti-
all the PRP studies, including questionable blinding, failure to inflammatory effects of fiber due to decreased adiposity, or bene-
conceal allocation, selective reporting, inappropriate statistical ficial changes in the microbiome. While these are associations from
analyses as well as heterogeneous patient populations, outcome observational cohorts and thus cannot prove causality, until the
measures and PRP preparation31. Although the “brave new world” definitive RCTs are performed, suggesting patients with knee OA
aspect of regenerative medicine is appealing, especially to patients adhere to the recommended average daily fiber intake of 25 g per
desperate for pain relief, the best quality reviews suggest we don't day is very low risk and may have significant benefits.
yet have strong enough data to support recommending these
therapies for our patients with OA. Comorbidities and OA

Opioids in OA Although physical therapy is known to benefit patients with OA,

clinicians may be hesitant to prescribe exercise therapy to patients
How to best treat the pain of OA, especially knee OA, remains a with OA and significant medical comorbidities, worried about the
major public health challenge, especially with the projected risk of adverse events. This would preclude a significant number of
increased rates of knee OA. Opioids are effective treatments for knee patients from receiving physical therapy, as between 30 and 50% of
OA pain, and given the known gastrointestinal and cardiovascular people with heart disease, diabetes, and obesity carry a doctor-
risks of nonsteroidal anti-inflammatories it has been argued there diagnosis of arthritis2. In addition, these patients may particularly
should be a role for opioid medication in the treatment algorithm of benefit from this therapy, as comorbidities are known to be associ-
OA33. However, there are little rigorous data to help guide treatment ated with worse pain and physical function in patients who have hip
decisions. Krebs et al. performed a pragmatic randomized trial and knee OA41. Rooji et al. performed a randomized controlled trial to
among Veterans Administration patients, to evaluate the best see whether exercise therapy can administered safely and effectively
strategy for managing chronic back pain or pain due to hip and knee to subjects with knee OA and significant clinical comorbidities42.
osteoarthritis34. Patients were randomized to a flexible treat-to- Subjects had knee OA as well as one of coronary heart disease, heart
target strategy of sequential opioid medications or sequential failure, type 2 diabetes, chronic obstructive pulmonary disease
non-opioid medications, and were followed for 12 months. The (COPD) or a body mass index of over 30 kg/m2. The comorbidity had
investigators found that the opioid medication strategy was not to interfere with daily activities, and patients had to be receiving
superior to the non-opioid approach for either pain-related function active treatment for the comorbidity. The intervention was a 20 week
or pain interference, and that there were significantly more adverse tailored exercise therapy program, which could be adapted to
events in the opioid group. Since only 35% of their subjects had hip accommodate specific comorbidities, and was administered by
or knee OA, the investigators were not powered to look at OA pa- trained physical therapists. Subjects were randomly assigned to
tients separately; however, a post hoc sensitivity analyses did not receive either the intervention immediately, or be placed on a
show any significant differences in outcomes between the back pain waiting list after which they would be eligible to receive physical
and OA groups. Although results from this Veterans Administration therapy. These investigators found that there were clinically and
cohort should be replicated in other populations, this well-done statistically significant improvements in physical function, which
trial suggests no clinical advantage to utilizing narcotics in the were maintained for 32 weeks, as well as a trend towards less pain.
treatment of painful hip or knee OA. Although underpowered to evaluate adverse events, there were no
Another study from the OAI also suggests there are increased serious adverse events reported. This is the first study to document
adverse events associated with using opioids35. Subjects with or at that careful administration of a tailored exercise program can be
high risk of developing knee OA had a 22% increased risk of falls administered safely and effectively to a high-risk knee OA popula-
compared to patients not receiving opioids. Opioids are also often tion, and that benefits last beyond program completion.
used for patients with severe knee OA waiting to undergo total knee
replacement. A recent cohort study suggests that patients with Conclusions
knee OA who use opioids for pain relief before their surgery have
less pain relief 6-months after TKR than those who do not use In conclusion, this review highlights some of the advances in
narcotics36. This is an important observation, as while most pa- clinical osteoarthritis published over the past 12 months. To ensure
tients have excellent results after TKR, multiple studies have shown ongoing progress, it is vital that innovative clinical investigators
that up to 30% of patients undergoing TKR have chronic pain continue to be encouraged and supported, to optimize quality of life
 L.A. Mandl / Osteoarthritis and Cartilage 27 (2019) 359e364 363

for the growing number of patients living with the pain and osteoarthritis: data from the osteoarthritis initiative. Clin
disability of osteoarthritis. Rheumatol 2017;36:1083e9.
13. Davis JE, Liu SH, Lapane K, Harkey MS, Price LL, Lu B, et al.
Author contributions Adults with incident accelerated knee osteoarthritis are more
likely to receive a knee replacement: data from the osteoar-
The author (LAM) was responsible for conception and design of thritis initiative. Clin Rheumatol 2018;37:1115e8.
the study, data review and interpretation, drafting and critical 14. McAlindon TE, LaValley MP, Harvey WF, Price LL, Driban JB,
revision of the article. Zhang M, et al. Effect of intra-articular triamcinolone vs saline
on knee cartilage volume and pain in patients with knee
Conflict of interest osteoarthritis: a randomized clinical trial. J Am Med Assoc
LAM is an associate editor at Annals of Internal Medicine for which 2017;317:1967e75.
she receives compensation. She also receives royalties from Wolters 15. Raynauld JP, Buckland-Wright C, Ward R, Choquett D,
Kluwer for contributing to Up-To-Date, the evidence-based online Haraoui B, Martel-Pelletier J, et al. Safety and efficacy of long-
clinical resource. term intraarticular steroid injections in osteoarthritis of the
knee: a randomized, double-blind, placebo-controlled trial.
Role of the funding source Arthritis Rheum 2003;48:370e7.
This work was not supported by any external funding. 16. Conaghan PG, Cohen SB, Berenbaum F, Lufkin J, Johnson JR,
Bodick N. Brief report: a phase IIb trial of a novel extended-
Acknowledgments release microsphere formulation of triamcinolone acetonide
for intraarticular injection in knee osteoarthritis. Arthritis
Thanks to Drs. David Hunter, Jeffrey Katz, Grace Lo and Tuhina Rheumatol 2018;70:204e11.
Neogi for their opinions regarding the relevance and clinical impact 17. Kingsbury SR, Tharmanathan P, Keding A, Arden NK, Birrell F,
of articles being considered for inclusion in this review. Cockayne S, et al. Hydroxychloroquine effectiveness in
reducing symptoms of hand osteoarthritis: a randomized trial.
References Ann Intern Med 2018;168:385e95.
18. Lee W, Ruijgrok L, Boxma-de Klerk B, Kok MR, Kloppenburg M,
1. Felson DT, Lawrence RC, Dieppe PA, Hirsch R, Helmick CG, Gerards A, et al. Efficacy of hydroxychloroquine in hand
Jordan JM, et al. Osteoarthritis: new insights. Part 1: the dis- osteoarthritis: a randomized, double blind, placebo-controlled
ease and its risk factors. Ann Intern Med 2000;133:635e46. trial. Arthritis Care Res (Hoboken) 2017;9:1320e5.
2. Prevalence of doctor-diagnosed arthritis and arthritis- 19. Aitken D, Laslett LL, Pan F, Haugen IK, Otahal P, Bellamy N,
attributable activity limitation e United States, 2010-2012. et al. A randomised double-blind placebo-controlled crossover
MMWR Morb Mortal Wkly Rep 2013;62:869e73. trial of HUMira (adalimumab) for erosive hand OsteoaRthritis
3. Michaud CM, McKenna MT, Begg S, Tomijima N, Majmudar M, e the HUMOR trial. Osteoarthr Cartil 2018;26:880e7.
Bulzacchelli MT, et al. The burden of disease and injury in the 20. Neogi T, Li S, Peloquin C, Misra D, Zhang Y. Effect of
United States 1996. Popul Health Metr 2006;4:11. bisphosphonates on knee replacement surgery. Ann Rheum
4. Ogden CL, Carroll MD, Lawman HG, Fryar CD, Kruszon- Dis 2018;77:92e7.
Moran D, Kit BK, et al. TRends in obesity prevalence among 21. Fu SH, Wang CY, Yang RS, Wu FL, Hsiao FY. Bisphosphonate
children and adolescents in the United States, 1988-1994 use and the risk of undergoing total knee arthroplasty in
through 2013-2014. J Am Med Assoc 2016;315:2292e9. osteoporotic patients with osteoarthritis: a nationwide cohort
5. Current Population Reports 2016. At: https://www.census.gov/ study in Taiwan. J Bone Jt Surg Am 2017;99:938e46.
prod/2014pubs/p25-1140.pdf; 2016 (Accessed 16 July 2018). 22. Vaysbrot EE, Osani MC, Musetti MC, McAlindon TE,
6. Lohmander LS, Englund PM, Dahl LL, Roos EM. The long-term Bannuru RR. Are bisphosphonates efficacious in knee osteo-
consequence of anterior cruciate ligament and meniscus in- arthritis? A meta-analysis of randomized controlled trials.
juries: osteoarthritis. Am J Sport Med 2007;35:1756e69. Osteoarthr Cartil 2018;26:154e64.
7. Flegal KM, Kruszon-Moran D, Carroll MD, Fryar CD, Ogden CL. 23. Saksena J, Platts AD, Dowd GS. Recurrent haemarthrosis
TRends in obesity among adults in the United States, 2005 to following total knee replacement. Knee 2010;17:7e14.
2014. J Am Med Assoc 2016;315:2284e91. 24. Felson DT, Niu J, Neogi T, Goggins J, Nevitt MC, Roemer F, et al.
8. Wallace IJ, Worthington S, Felson DT, Jurmain RD, Wren KT, Synovitis and the risk of knee osteoarthritis: the MOST Study.
Maijanen H, et al. Knee osteoarthritis has doubled in preva- Osteoarthr Cartil 2016;24:458e64.
lence since the mid-20th century. Proc Natl Acad Sci U S A 25. Okuno Y, Korchi AM, Shinjo T, Kato S, Kaneko T. Midterm
2017;114:9332. clinical outcomes and MR Imaging changes after transcatheter
9. Sharma L, Hochberg M, Nevitt M, Guermazi A, Roemer F, arterial embolization as a treatment for mild to moderate
Crema MD, et al. Knee tissue lesions and prediction of incident radiographic knee osteoarthritis resistant to conservative
knee osteoarthritis over 7 years in a cohort of persons at treatment. J Vasc Interv Radiol 2017;28:995e1002.
higher risk. Osteoarthr Cartil 2017;25:1068e75. 26. Allen KD, Arbeeva L, Callahan LF, Golightly YM, Goode AP,
10. Lankhorst NE, Damen J, Oei EH, Verhaar JAN, Kloppenburg M, Heiderscheit BC, et al. Physical therapy vs internet-based ex-
Bierma-Zeinstra SMA, et al. Incidence, prevalence, natural ercise training for patients with knee osteoarthritis: results of a
course and prognosis of patellofemoral osteoarthritis: the randomized controlled trial. Osteoarthr Cartil 2018;26:383e96.
Cohort Hip and Cohort Knee study. Osteoarthr Cartil 2017;25: 27. O'Moore KA, Newby JM, Andrews G, Hunter DJ, Bennell K,
647e53. Smith J, et al. Internet cognitive-behavioral therapy for depres-
11. Mills K, Hunter DJ. Patellofemoral joint osteoarthritis: an sion in older adults with knee osteoarthritis: a randomized
individualised pathomechanical approach to management. controlled trial. Arthritis Care Res (Hoboken) 2018;70:61e70.
Best Pract Res Clin Rheumatol 2014;28:73e91. 28. Prochaska JO, DiClemente CC. Stages of change in the modifi-
12. Davis J, Eaton CB, Lo GH, Lu B, Price LL, McAlindon TE, et al. cation of problem behaviors. Prog Behav Modif 1992;28:
Knee symptoms among adults at risk for accelerated knee 183e218.
364 L.A. Mandl / Osteoarthritis and Cartilage 27 (2019) 359e364

29. Xing D, Wang Q. Mesenchymal Stem Cells Injections for Knee 36. Smith SR, Bido J, Collins JE, Yang H, Katz JN, Losina E. Impact of
Osteoarthritis: A Systematic Overview 2017. preoperative opioid use on total knee arthroplasty outcomes.
30. Pas HI, Winters M, Haisma HJ, Koenis MJ, Tol JL, Moen MH. JBJS 2017;99:803e8.
Stem cell injections in knee osteoarthritis: a systematic review 37. Wylde V, Hewlett S, Learmonth ID, Dieppe P. Persistent pain
of the literature. Br J Sport Med 2017;51:1125e33. after joint replacement: prevalence, sensory qualities, and
31. Bennell KL, Hunter DJ, Paterson KL. Platelet-rich plasma for the postoperative determinants. Pain 2011;152:566e72.
management of hip and knee osteoarthritis. Curr Rheumatol 39. Dai Z, Lu N, Niu J, Felson DT, Zhang Y. Dietary fiber intake in
Rep 2017;19:24. relation to knee pain trajectory. Arthritis Care Res (Hoboken)
32. Beswick AD, Wylde V, Gooberman-Hill R, Blom A, Dieppe P. 2017;69:1331e9.
What proportion of patients report long-term pain after total 40. Dai Z, Niu J, Zhang Y, Jacques P, Felson DT. Dietary intake of
hip or knee replacement for osteoarthritis? A systematic re- fibre and risk of knee osteoarthritis in two US prospective
view of prospective studies in unselected patients. BMJ open cohorts. Ann Rheum Dis 2017;76:1411e9.
2012;2:e000435. 41. Calders P, Van Ginckel A. Presence of comorbidities and
33. Smith SR, Deshpande BR, Collins JE, Katz JN, Losina E. prognosis of clinical symptoms in knee and/or hip osteoar-
Comparative pain reduction of oral non-steroidal anti-in- thritis: a systematic review and meta-analysis. Semin Arthritis
flammatory drugs and opioids for knee osteoarthritis: sys- Rheum 2018;47:805e13.
tematic analytic review. Osteoarthr Cartil 2016;24:962e72. 42. de Rooij M, van der Leeden M, Cheung J, van der Esch M,
34. Krebs EE, Gravely A, Nugent S, Jensen AC, DeRonne B, Hakkinen A, Haverkamp D, et al. Efficacy of tailored exercise
Goldsmith ES, et al. Effect of opioid vs nonopioid medications therapy on physical functioning in patients with knee osteo-
on pain-related function in patients with chronic back pain or arthritis and comorbidity: a randomized controlled trial.
hip or knee osteoarthritis pain: the SPACE randomized clinical Arthritis Care Res (Hoboken) 2017;69:807e16.
trial. Jama 2018;319:872e82. 43. Collins JE, Losina E, Nevitt MC, Roemer FW, Guermazi A,
35. Lo-Ciganic WH, Floden L, Lee JK, Ashbeck EL, Zhou L, Lynch JA, et al. Semiquantitative imaging biomarkers of knee
Chinthammit C, et al. Analgesic use and risk of recurrent falls osteoarthritis progression: data from the foundation for the
in participants with or at risk of knee osteoarthritis: data from national institutes of health osteoarthritis biomarkers con-
the osteoarthritis initiative. Osteoarthr Cartil 2017;25:1390e8. sortium. Arthritis Rheumatol 2016;68(10):2422e31.