MedicineLakex

An algorithm recommendation for the management of knee osteoarthritis in europe and internationally_ a report from a task force of the european society for clinical and economic aspects of osteoporosis and osteoarthritis (esceo)

Contents lists available at Seminars in Arthritis and Rheumatism journal homepage: An algorithm recommendation for the management of kneeosteoarthritis in Europe and internationally: A report from atask force of the European Society for Clinical and EconomicAspects of Osteoporosis and Osteoarthritis (ESCEO) Olivier Bruyère, n, Cyrus Cooper, MD, PhDJean-Pierre Pelletier, MD, Jaime Branco, MD, Maria Luisa Brandi, MDFrancis Guillemin, MD, Marc C. Hochberg, MD, John A. Kanis, Tore K. Kvien, MD, Johanne Martel-Pelletier, PhD, René Rizzoli, MD, Stuart Silverman, MD,Jean-Yves Reginster, MD, PhD a Support Unit in Epidemiology and Biostatistics, Department of Public Health, Epidemiology and Health Economics, University of Liège,CHU Sart Tilman, 4000 Liège, Belgiumb MRC Lifecourse Epidemiology Unit, University of Southampton, Southampton, UKc NIHR Musculoskeletal Biomedical Research Unit, University of Oxford, Oxford, UKd Osteoarthritis Research Unit, University of Montreal Hospital Research Centre (CRCHUM), Montreal, Quebec, Canadae CEDOC, Department of Rheumatology, Faculdade de Ciências Médicas, Universidade Nova de Lisboa/CHLO, EPE—Hospital Egas Moniz,Lisbon, Portugalf Department of Internal Medicine, University of Florence, Florence, Italyg Université de Lorraine, Université Paris Descartes, Nancy, Franceh Division of Rheumatology & Clinical Immunology, Department of Medicine, University of Maryland School of Medicine, Baltimore, MDi Geriatric Research, Education and Clinical Center, Baltimore, MDj Health Care System, Baltimore, MDk WHO Collaborating Centre for Metabolic Bone Diseases, University of Sheffield, Sheffield, UKl Department of Rheumatology, Diakonhjemmet Hospital, Oslo, Norwaym Service of Bone Diseases, Geneva University Hospitals and Faculty of Medicine, Geneva, Switzerlandn Cedars-Sinai Medical Center, Los Angeles, CAo OMC Clinical Research Center, Beverly Hills, CA Objectives: Existing practice guidelines for osteoarthritis (OA) analyze the evidence behind eachproposed treatment but do not prioritize the interventions in a given sequence. The objective was to develop a treatment algorithm recommendation that is easier to interpret for the prescribing physician based on the available evidence and that is applicable in Europe and internationally. The knee was usedas the model OA joint.
Methods: ESCEO assembled a task force of 13 international experts (rheumatologists, clinical epidemi-ologists, and clinical scientists). Existing guidelines were reviewed; all interventions listed and recentevidence were retrieved using established databases. A first schematic flow chart with treatmentprioritization was discussed in a 1-day meeting and shaped to the treatment algorithm. Fine-tuningoccurred by electronic communication and three consultation rounds until consensus.
Results: Basic principles consist of the need for a combined pharmacological and non-pharmacologicaltreatment with a core set of initial measures, including information access/education, weight loss ifoverweight, and an appropriate exercise program. Four multimodal steps are then established. Step1 consists of background therapy, either non-pharmacological (referral to a physical therapist for re-alignment treatment if needed and sequential introduction of further physical interventions initially andat any time thereafter) or pharmacological. The latter consists of chronic Symptomatic Slow-Acting Drugsfor OA (e.g., prescription glucosamine sulfate and/or chondroitin sulfate) with paracetamol at-need;topical NSAIDs are added in the still symptomatic patient. Step 2 consists of the advanced pharmaco-logical management in the persistent symptomatic patient and is centered on the use of oral COX-2 n Corresponding author.
0049-0172/& 2014 The Authors. Published by Elsevier HS Journals, Inc. This is an open access article under the CC BY-NC-ND license(http://creativecommons.org/licenses/by-nc-nd/3.0/).
O. Bruyère et al. / Seminars in Arthritis and Rheumatism ] (2014) ]]]–]]] selective or non-selective NSAIDs, chosen based on concomitant risk factors, with intra-articularcorticosteroids or hyaluronate for further symptom relief if insufficient. In Step 3, the last pharmaco-logical attempts before surgery are represented by weak opioids and other central analgesics. Finally,Step 4 consists of end-stage disease management and surgery, with classical opioids as a difficult-to-manage alternative when surgery is contraindicated.
Conclusions: The proposed treatment algorithm may represent a new framework for the development offuture guidelines for the management of OA, more easily accessible to physicians.
& 2014 The Authors. Published by Elsevier HS Journals, Inc. This is an open access article under the CC BY- most recent drafts released for consultation and circu-late them to the task force. Afterwards, an initial list of all Osteoarthritis (OA) is the most common form of arthritis and a interventions considered in the different documents was com- major cause of disability The most prevalent OA localization is piled, preparing a first and very schematic flow chart for possible the knee joint, and symptomatic knee OA affects 24% of the general prioritization of the different interventions, to the extent possible population ; thus, it represents the model for the development based on the information within the documents. For each inter- of treatment guidelines. Recommendations for the management of vention, a complete literature search was performed to identify knee OA have been issued by national, continental, or global new or additional randomized controlled trials and systematic scientific authorities, including, among others, the European reviews/meta-analyses, if any, not used in the existing guidelines.
League Against Rheumatism (EULAR) ; the American College The MEDLINE (PubMed) database was searched using each inter- of Rheumatology (ACR) ; and the Osteoarthritis Research vention as a search term together with "osteoarthritis," limiting Society International (OARSI) The UK National Institute for results to " humans," "randomized controlled trial," "meta-analy- Health and Clinical Excellence (NICE) has also issued a guideline sis," "review" and "systematic review," and "guideline." A similar document for the management of OA in adults search was adapted for the Embase database and each interven- The European Society for Clinical and Economic Aspects of tion was also searched in the Cochrane Central Register of Con- Osteoporosis and Osteoarthritis (ESCEO) has promoted the crea- trolled Trials, the Cochrane Database of Systematic Reviews, and tion of different expert working groups in the field of OA to the Centre for Reviews and Dissemination of the University of analyze and generate consensus documents on disease manage- York. The reference list of relevant retrieved articles was hand- ment . With respect to treatment guidelines, ESCEO felt the searched for additional resources, while members of the task force need to find a common denominator to the published guidelines were interrogated for their knowledge on articles or congress and recommendation documents from different sources and gen- abstracts in press. A free web search was also performed and erate a treatment algorithm applicable throughout Europe and considered. Searches were performed from the year 2000 and elsewhere. Most of the existing practice guidelines analyze the updated until February 2014, with the additional evidence con- evidence behind each proposed treatment but do not (often stantly provided to the task force members for selection of the best deliberately) prioritize the interventions because few clinical trials evidence according to the panel.
have been designed to study the effect of a given treatment in Initial material was presented to the task force, together patients in whom initial therapies have failed and/or when and with the first schematic draft flow chart, in a 1-day meeting how new treatments should be introduced. Therefore, treatment at the end of September 2013. During a detailed discussion, the remains based on the individualized assessment of the patient, initial flow chart was shaped to the treatment algorithm. Each step taking into account patients' needs and preferences, or the sub- and intervention within the steps were discussed and amended jective interpretation of the evidence by the physician. On the until consensus agreement was achieved within the task force.
other hand, careful analysis of the evidence actually allows one to A first draft of the final algorithm and the supporting evidence prioritize interventions and guide physicians into progressive and was then prepared (O.B.); three rounds of consultation occurred logical steps. In addition, it was necessary to synthesize the by electronic communication between December 2013 and recommendations for the European situation, adapting global February 2014 on the draft manuscript: all corrections and guidelines to it or using what is applicable from other suggestions by each member were shared with the rest of the continental guidelines and broadening the perspective of some task force and included till final consensus with the third national guidelines providing an initial update of current consultation round.
European guidelines that can be used virtually universally.
ESCEO is a not-for-profit organization dedicated to providing Finally, the present effort may represent a new framework for the practitioners with the latest clinical evidence in the field of bone, development of recommendations for the treatment of OA that is joint, and muscle disorders: the development of practice recom- more practical and intuitive for practicing physicians than a mere mendations is within the institutional scope of the organization to exposition of the available and often contradicting evidence.
allow specialist and non-specialist practitioners to organizetheir daily clinical activity in an evidence-based medicine per-spective, with a cost-conscious perception. ESCEO receives Unre- companies that do not influence the institutional activities of the ESCEO gathered an international task force of 13 members, organization. In particular, the present recommendations were consisting of 11 rheumatologists (8 from Europe, 2 from USA, and developed independently of any of the funding sources that had no 1 from Canada), 1 clinical epidemiologist, and 1 clinical scientist, role in the decision to prepare this document and its implementa- all of whom are experienced in the performance, analysis, and tion, revisions, and approval for publication. In addition, each interpretation of the clinical trial evidence related to OA.
member of the task force individually agreed to declare their One member of the task force (O.B.) was entrusted with the potential conflict of interest, if any, in the process of article task to collect the principal guidelines considered, including the O. Bruyère et al. / Seminars in Arthritis and Rheumatism ] (2014) ]]]–]]] been made and with disease progression. In agreement with thebasic principle of treatment recommendation, which requires Basic principles and core set for a treatment algorithm in knee OA combination of treatment modalities, parallel addition of sequen-tial non-pharmacological and pharmacological therapies should be Appropriate diagnosis of knee OA is an essential pre-requisite established. The consequent treatment algorithm that was derived to treatment; in this respect, EULAR recently published updated from the present effort is depicted in the and can be recommendations .
described as follows.
Virtually all existing practice guidelines agree that combination of treatment modalities, including non-pharmacological and phar- Step 1: Background treatment macological intervention, is strongly recommended. It was the During Step 1, which follows the core set, further background firm opinion of the task force that this basic principle is valid and physical remedies should be established as needed. In parallel, and makes the effort to develop a treatment algorithm an absolute if the patient is still symptomatic, background pharmacological priority in order to prevent physicians from being confused as to therapy should be started and progressively moved toward com- how treatments should be prioritized and possibly added on for bination treatment as soon as the clinical response is not The core set proposed by NICE was adopted and expanded by the task force to represent the initial measures and interventions Step 1-a: Non-pharmacological background treatment. During Step that every patient with knee OA should undergo. In particular, 1, the patient should be referred to a physical therapist forassessment to determine whether physical treatments should be (1) Information access and education consists in providing to the introduced. In particular, the physical therapist should first patient the necessary knowledge about the nature of evaluate whether correction for malalignment is necessary.
the disease and the objectives of treatment. If necessary, the Moreover, he/she should also assess, during Step 1 and physician should prompt changes in the patient's lifestyle throughout steps thereafter at any time, whether other physical toward behaviors that may have a beneficial impact on joint measures may be useful for additional symptom relief in parallel to protection or at least not worsen the progression of the disease the pharmacological interventions established by the physician.
or of its symptoms. It is recognized that these measures have Varus or valgus malalignment is a risk factor for knee OA and its minimal effect on OA symptoms, but they are essential for progression. Therefore, there is a theoretical rationale for using treatment adherence. EULAR has recently published compre- biomechanical interventions such as braces or insoles in patients hensive recommendations for the non-pharmacological man- with unicompartmental tibiofemoral OA to reduce malalignment, agement of hip and knee osteoarthritis that provided extensive to reduce the consequent articular stress, and thus to improve pain guidance on the principles of information and education, as and function, or even retard disease progression. Despite signifi- well as of lifestyle changes cant heterogeneity and poor trial quality, there is reasonable (2) Weight loss if overweight. Analysis of the available evidence evidence to suggest that knee braces actually improve biomechan- indicates that at least 5% weight loss within 6 months induces ical imbalance and may improve knee OA symptoms The a small but well-substantiated symptomatic benefit, more evi- same may apply with multi-modal re-alignment treatment that dent on physical function than on pain, where the effect is less includes braces, foot orthoses, and appropriate footwear predictable The task force strongly felt that a threshold Among foot orthoses, there are many studies on laterally wedged should be indicated and, based on previous evidence and a insoles in medial compartment knee OA aimed to increase foot recent high-quality trial , weight loss should be targeted to at pronation, thereby reducing medial compartment loading. Their least 10% to achieve significant symptom benefit. A similar degree biomechanical effects are more inconsistent and the clinical of weight loss has also been indicated to improve the quality and efficacy is controversial; in particular, a recent high-quality meta- thickness of medial femoral compartment cartilage analysis found a significant effect on pain in all randomized (3) Exercise program. Education should include information about controlled studies combined, but no effect when the more reliable exercise and physical activity since exercise produces higher quality trials were considered . There are fewer studies benefit on both pain and function in patients with knee on laterally wedged insoles with subtalar strapping and on osteoarthritis by different delivery modes (individual, group- medially wedged insoles for unicompartmental lateral OA, based, or home programs) Although the optimal exercise although they tend to show efficacy . There is insufficient dosing and rate of progression in its application remains evidence to determine whether braces or insoles affect the pro- unclear, expert opinion suggests that the intensity and/or gression of knee OA. EULAR did not recommend the use of insoles duration of exercise should be increased over time There among non-pharmacological treatments for knee OA, also in view is good evidence that water-based exercise is effective on both of an increased risk of adverse effects On the other hand, pain and function . However, specific quadriceps strength- appropriate footwear (including shock-absorbing soles, foot arches ening exercises or strength training for the lower limb, support, and control for foot pronation if necessary, while avoiding together with aerobic training such as walking, remain the raised heels) should be recommended best documented exercise approaches experts suggest Ideal patients for bracing are younger individuals, more physi- that mixed programs (to include muscle strengthening, aero- cally active, not severely obese, with unicompartmental sympto- bic capacity, and flexibility/range of motion) should be recom- matic tibiofemoral OA and malalignment that is reducible by mended as long as minimal intensity requirements are valgus or varus stress maneuvers on physical examination. For met . Recent evidence suggests that tai chi is also effective laterally wedged insoles, patients should also probably have early in relieving symptoms and mild disease.
Among additional physical remedies, access to walking aids is an important help in providing security to patients: although OA treatment algorithm: Beyond the core set appropriate clinical studies are scarce, a recent randomized trialconfirmed that use of a cane improves symptoms in knee OA It is a common clinical experience that core therapies are Referral to a physical therapist may also include assessment for the usually insufficient to fully control symptoms after diagnosis has use of thermal therapies, such as ultrasound, for which there is O. Bruyère et al. / Seminars in Arthritis and Rheumatism ] (2014) ]]]–]]] BASIC PRINCIPLE AND CORE SET
Combination of treatment modalities, including non-pharmacological and pharmacological
therapies is strongly recommended
Core set: - Information/Education
- Weight loss if overweight
- Exercise program (aerobic, strengthening)
STEP 1: Background treatment
Referral to physical therapist for:
if needed
(to control malalignment)
- (Paracetamol on a regular basis) - Chronic SYSADOA: prescripon glucosamine sulfate and/or chondroin sulfate ± as needed paracetamol if sll symptomac ADD
if symptomac ADDITION at any me
- Topical NSAIDs - (Topical capsaicin) Manual therapyPatellar tapingChinese acupunctureTENS STEP 2: Advanced pharmacological management in the persistent symptomatic patient
if sll or severely symptomac
- Intermient or connuous (longer cycles) oral NSAIDs INCREASED GI RISK* INCREASED CV RISK INCREASED RENAL RISK - Non selecve NSAID - Cox-2 selecve NSAID - Prefer naproxen - Avoid NSAIDs† - Avoid high-dose diclofenac and - Cox-2 selecve NSAID - Avoid non-selecve NSAIDs ibuprofen (if on low-dose aspirin) - Cauon with other non-selecve NSAIDs - Avoid Cox-2 selecve NSAIDs *Including use of low dose aspirin †With glomerular filtraon rate <30 cc/min; cauon in other cases if sll symptomac
- Intraarcular hyaluronate - Intraarcular corcosteroids STEP 3: Last pharmacological attempts
- Short-term weak opioids STEP 4: End-stage disease management and surgery
if severely symptomac and poor quality of life
- Total joint replacement - (Unicompartmental knee replacement) - Opioid analgesics Fig. Knee osteoarthritis treatment algorithm. COX-2, cyclooxygenase-2; CV, cardiovascular; GI, gastrointestinal; NSAIDs, non-steroidal anti-inflammatory drugs; PPI, protonpump inhibitor; SYSADOA, Symptomatic Slow Acting Drugs in Osteoarthritis.
some evidence of efficacy, albeit in low-quality studies .
manual therapy in combination with exercise, or patellar taping Indeed, more recent randomized trials failed to show additional where needed. Balneotherapy and, especially, acupuncture ranked improvement of ultrasound over a sham procedure , or for high in a recent network meta-analysis of all physical treatments O. Bruyère et al. / Seminars in Arthritis and Rheumatism ] (2014) ]]]–]]] for knee OA On the other hand, other studies suggest that the been evaluated by the FDA In addition, such dietary supple- benefit of acupuncture is small, especially compared with sham ments often contain glucosamine hydrochloride and a large, NIH- acupuncture , and possibly due to expectation or placebo sponsored trial showed no benefit of this preparation similar effect Finally, Transcutaneous Electric Nerve Stimulation to other studies of glucosamine hydrochloride ,; indeed, these (TENS) is also a possible physical intervention; the evidence is formulations have a different pharmacokinetic profile than pre- scarce, especially because of the availability (as it often happens scription glucosamine sulfate 1500 mg once daily approved in for physical remedies) of mainly small and low-quality studies Europe and appropriate bioequivalence studies should be . However, a recent larger randomized trial suggested that conducted before recommending glucosamine generic or OTC TENS may reduce the need for analgesic medications . Inter- products . Long-term prescription glucosamine sulfate may estingly, the recent ACR guidelines suggested an approach that delay joint structure changes suggesting potential benefit would advise reserving acupuncture and TENS as a non- beyond symptom control when used early in the management of pharmacological alternative to surgery when this is contraindi- knee OA as recommended here.
cated or the patient is unwilling to undergo surgery. There are no Chondroitin sulfate may offer similar benefits on joint structure specific studies in this respect and thus it is correct that both changes in patients with mild-to-moderate disease using acupuncture and TENS are listed here and considered throughout prescription chondroitin 4&6 sulfate. It should, however, be the algorithm flow including, but not necessarily, as an alternative acknowledged that joint structure modification is not an approved indication for either prescription chondroitin or glucosaminesulfate but a potential benefit when they are used for long-term Step 1-b: Pharmacological background treatment. The aim of Step symptom control. In this respect, the data on chondroitin sulfate 1 pharmacological treatment is to establish a first chronic therapy have been reported as conflicting but the effect size on pain that may improve or control symptoms or at least provide rescue in meta-analyses ranges from 0.13 (0.00–0.27) to 0.75 (0.50–0.99) On the other hand, full data of one pivotal trial on chondroitin Paracetamol at doses no greater than 3 g/day on a regular basis 4&6 sulfate have been published more recently and show chon- is recommended as an initial pharmacological approach in most droitin sulfate to reduce joint structural changes with a sympto- clinical guidelines, despite its minimal effects on symptoms but in matic effect that could be clinically relevant as confirmed in the presumption of acceptable safety and affordable price. How- another recent study The prescription drug used in this study ever, a recent meta-analysis outlined that the vast majority of should be distinguished from low-quality OTC products clinical trials (of either immediate release or extended release available outside of Europe. As for glucosamine sulfate, the formulations) were performed over durations of less than negative conclusions of a recent network meta-analysis were 6 months, thus questioning paracetamol chronic treatment role; withdrawn by the editor indeed, the only placebo-controlled study of 6-month duration Beside their efficacy record, both glucosamine sulfate and found a significant effect on function but not on pain This chondroitin are safe medications, with no difference in adverse meta-analysis also confirmed the small effect size, less than 0.20, effects compared with placebo , which would also detected in a previous OARSI guideline update and confirmed strengthen their role as chronic background treatments. Glucos- in large trials or those analyzed by intention-to-treat . In amine and chondroitin sulfate are often used in combination as addition, there is accumulating evidence for an increased risk of dietary supplements. Unfortunately, there are no published trials gastrointestinal adverse events with paracetamol use, with sig- of the two pharmaceutical-grade prescription preparations com- nificant elevation in liver enzymes . Indeed, paracetamol is the bined. On the other hand, the previously mentioned NIH- most frequent cause of drug-induced liver injury in the United sponsored trial while finding no overall benefit, described a States and over half of such cases are due to unintentional significant symptomatic effect in an exploratory analysis of patients with moderate-to-severe pain receiving a glucosamine A safer and more sensible approach would be to use as a hydrochloride and chondroitin sulfate combination Indeed, a background therapy, chronic Symptomatic Slow-Acting Drugs for similar combination has been shown to have comparable efficacy Osteoarthritis (SYSADOAs), with as needed paracetamol for short- to celecoxib after 6 months in knee OA patients with moderate-to- term, rescue analgesia. Among SYSADOAs, prescription glucos- severe pain (Hochberg et al., abstract presented at OARSI 2014).
amine sulfate should be differentiated from other glucosamine More interestingly, and following preliminary data from the preparations. The latest update of a specific Cochrane Review Osteoarthritis Initiative database , a recently published trial clearly showed that while the overall pain benefit of all glucos- from Australia found significant and clinically relevant joint amine formulations together is jeopardized by high heterogeneity, structure modification after 2 years with a non-prescription or subgroup analysis of non-prescription glucosamine high-quality pharmaceutical-grade chondroitin/glucosamine sulfate combina- trials showed no benefit versus placebo Conversely, high- tion, with apparently no effects on symptoms given the very mild quality trials of the patented prescription formulation approved in and placebo responsive population studied .
Europe and elsewhere (crystalline glucosamine sulfate) showed The evidence for efficacy of other SYSADOAs, such as avocado that it was superior to placebo in the treatment of pain and soybean unsaponifiables (ASU) and diacerein, is more scarce functional impairment all three pivotal trials are . With respect to diacerein, the European Medicines Agency long-term studies of 6 months to 3 years treatment in patients (EMA) recently informed that despite the most recent meta- with mild-to-moderate pain and the calculated global effect size, analysis showing some degree of efficacy without heterogeneity, without heterogeneity, is 0.27 (95% CI: 0.12–0.43) on pain and 0.33 safety issues may outline a negative benefit/risk ratio and a re- (95% CI: 0.17–0.48) on function , i.e., in about the same range of assessment was recommended Furthermore, newer drugs are short-term trials of oral non-steroidal anti-inflammatory drugs gaining credit for their possible role as disease-modifying agents in (NSAIDs) . A recent network meta-analysis, again mixing all OA, e.g., strontium ranelate (SR), a chemical entity currently formulations was criticized for severe methodological flaws marketed for the treatment of postmenopausal osteoporosis and and the journal editor withdrew the article's negative con- osteoporosis in males. SR exhibited effects on subchondral bone clusions . The ACR guidelines did not recommend glucosamine and cartilage, which may be suggestive of a beneficial effect on OA because they noted that glucosamine is not a prescription drug in progression. A recent high-quality 3-year placebo-controlled trial the USA, but only dietary supplements exist whose quality has not showed that SR decreases radiological progression of knee OA, O. Bruyère et al. / Seminars in Arthritis and Rheumatism ] (2014) ]]]–]]] together with improving symptoms The drug was well COX-2 selective, partially selective, or non-selective NSAIDs .
tolerated in this trial, but the EMA recently restricted the use of Drug choice within available NSAIDs is therefore dictated by their SR to the treatment of severe osteoporosis due to a possible safety profile according to the different risk factors and patients' increase in cardiovascular risk ; therefore, the role of the drug concomitant diseases and medical conditions.
in the management of OA will need further reassessment.
While COX-2-selective agents are associated with fewer gastro- If the patient is still symptomatic after establishing appropriate intestinal ulcer complications than non-selective NSAIDs over background pharmacological therapy with SYSADOAs and rescue short-term use, this is not clear over longer periods of treatment analgesia with paracetamol is insufficient, topical NSAIDs may be , in particular for celecoxib or etoricoxib , which are added to the pharmacological background and in parallel with the only COX-2-selective agents left in the market in Europe.
non-pharmacological treatment. The efficacy of topical NSAIDs Indeed, recent evidence suggests that coxibs significantly increase was established in several randomized trials that were systemati- the risk of upper gastrointestinal complications compared cally reviewed in 2004 the evidence shows a moderate effect to placebo, although at a lower rate than non-selective NSAIDs size in the relief of OA pain with high heterogeneity possibly . Thus, on one hand non-selective NSAIDs should always be due to differences between topical products. Recent randomized co-prescribed with proton pump inhibitors (PPI). However, for trials confirmed that there is no significant difference between the reasons described above and for recent cost-effectiveness topical NSAIDs and their oral counterpart, with topical formula- evidence the task force believes that in patients with normal tions showing better gastrointestinal safety but more local cuta- gastrointestinal risk, physicians should consider whether a PPI neous reactions, including in large studies and as discussed should be co-prescribed also with COX-2-selective NSAIDs. In in a comprehensive systematic review . However, most of the patients with increased gastrointestinal risk, non-selective NSAIDs studies were for r12 weeks, which would confirm topical NSAIDs' should be avoided and COX-2-selective NSAIDs should be co- role as cyclic add-on analgesics if the background treatment leaves prescribed with a PPI Concomitant use of aspirin increases the patient symptomatic. Only one study performed in general non-selective NSAID gastrointestinal risk and at least partially practice confirmed similar findings over 1 year, but the study negates COX-2-selective NSAID improved gastrointestinal toler- did not meet all quality standards of a rigorous randomized ance : addition of a PPI is beneficial and reduces the risk in controlled trial and was not blinded .
While topical rubefacients containing salicylates did not show There is little doubt that all NSAIDs, selective and non-selective, efficacy compared with placebo in OA, with increased risk of local increase the risk of serious cardiovascular events; however, adverse reactions there is no particular reason to recommend naproxen is usually associated with lower risk for thrombotic topical capsaicin instead of topical NSAIDs when all the available cardiovascular events . This was confirmed in a recent meta- evidence is reviewed . In fact, despite some evidence of analysis of individual patient data from 639 randomized controlled efficacy against placebo, there is increased risk of local adverse trials in which coxibs and diclofenac were shown to increase reactions and no comparative trials with topical or oral NSAIDs.
major vascular events by a third, mainly because of increasedmajor coronary events (including fatal and non-fatal myocardialinfarction) that occurred also with ibuprofen but not with nap- Step 2: Advanced pharmacological management in the persistent roxen that did not increase the general vascular risk, probably due symptomatic patients to its sustained suppression of platelet aggregation . In 2012, The task force agreed that patients with mild-to-moderate pain EMA stated that data suggest that naproxen may be associated due to knee OA can be appropriately managed by careful applica- with a lower risk for arterial thrombotic events than COX-2 tion of core set and Step 1 treatment modalities. However, if Step 1 inhibitors and other NSAIDs, although a small risk cannot be shows inadequate efficacy or in patients arriving to the observa- excluded . This latter opinion is shared by the FDA, which tion with moderate-to-severe pain, benefit may be achieved with recently decided not to recommend changes in the cardiovascular advanced pharmacological treatment. However, these treatments warning for naproxen based on the totality of the evidence from are less manageable and/or prone to more severe adverse systematic reviews of observational studies and randomized con- trolled trials (Hochberg MC, personal communication).
In this respect, a central role within the pharmacological While earlier evidence suggested that the relative risk of management of OA is traditionally represented by the use of oral cardiovascular events increases with increased baseline cardiovas- NSAIDs. Recourse to oral NSAIDs may occur early for rescue cular risk at the start of treatment with COX-2-selective NSAIDs analgesia in very short treatment cycles. In fact, oral non- , the most recent and comprehensive meta-analysis found that selective or COX-2-selective NSAIDs provide better symptom relief the proportional increase in risk was similar irrespective of base- than paracetamol with an effect size on pain of 0.29 (0.22– line risk ; however, the excess absolute risk is in any case 0.35) , which is double the effect of paracetamol . Indeed, higher in patients at high risk of major cardiovascular events and patients show a higher preference for NSAIDs compared to receiving coxibs or diclofenac and, probably, ibuprofen For paracetamol . Although systematic reviews of head-to- this reason, the task force suggests to avoid coxibs and high-dose head trials show no clear difference between glucosamine diclofenac or ibuprofen in high cardiovascular risk patients. How- sulfate and oral NSAIDs for pain or function , with better ever, this may apply to other non-selective NSAIDs as well safety of the former , the task force believes that based on (naproxen being a possible exception), based on observational the available clinical trial evidence, NSAIDs may be appropriate studies where the risk is persistently elevated suggesting in patients with more severe pain, particularly if the SYSADOA much caution. On the other hand, naproxen, because of its possibly has failed to effectively control symptoms. On the other hand, lower cardiovascular risk may be the preferred agent if an SYSADOAs may decrease NSAID use if adopted as a background NSAID should be used in patients at high cardiovascular risk.
Concomitant use of low-dose aspirin did not attenuate the Oral NSAIDs may be used intermittently or continuously, increased cardiovascular risk associated with COX-2-selective although even the latter is not represented by chronic use but by NSAIDs in placebo-controlled trials while observational longer cycles, because of safety concerns and the paucity of long- studies did not find increased cardiovascular risk when non- term trials. Recent systematic reviews of the many, mostly short- selective NSAIDs, other than ibuprofen, are added to low-dose term, clinical trials show no clear differences in efficacy between aspirin Actually, this does not apply to ibuprofen, which has a O. Bruyère et al. / Seminars in Arthritis and Rheumatism ] (2014) ]]]–]]] clinically relevant pharmacodynamic interaction with the actions Antidepressants are commonly used in chronic pain syndromes of aspirin and should be therefore avoided because they alter pain neurotransmitters (i.e., serotonin and The association of NSAIDs with increased cardiovascular risk is norepinephrine) centrally, which may to some extent be applicable a complex issue that, besides being directly mediated by prosta- also to tramadol. Indeed, central sensitization may play a role in noid imbalance may have indirect causality linked to the the severity of osteoarthritis pain . In this respect, evidence for increase in blood pressure, congestive heart failure, and renal the use of duloxetine in knee osteoarthritis derives mainly from dysfunction associated with all NSAIDs that should be always two randomized, placebo-controlled trials of short-term, 13 weeks, considered by physicians. In particular, it may be useful to suggest duration showing benefit in clinically relevant outcomes but a avoiding oral NSAID use in patients with increased renal risk, such similar likelihood to elicit adverse reactions, such as nausea, as chronic kidney disease with estimated glomerular filtration rate fatigue, constipation, dry mouth, and others . In a similarly short-term study, duloxetine improved knee pain in patients with In case of contraindications to NSAIDs, or if the patient is still an inadequate response to NSAIDs , which would confirm the symptomatic despite use of NSAIDs or was severely symptomatic, drug's possible role in advanced patient management. Additional intra-articular treatment may be applied.
and longer term evidence is needed before moving duloxetine to The role of intra-articular hyaluronic acid injections has been earlier steps in the management of the disease. Actually, early controversial, but most meta-analyses have shown a significant physical examination may allow recognition of patients with benefit in knee osteoarthritis, which is small when restricted to central sensitization who thus may benefit from central high-quality trials However, a more recent meta-analysis showed analgesics earlier in the course of the disease management.
an effect size of 0.34 (0.22–0.46), which is hindered by highheterogeneity but survives several sensitivity analyses based on trial Step 4: End-stage disease management and surgery quality and, importantly, the small effects last up to 6 months after Full review and advice on surgical procedures for the manage- treatment These favorable results are confirmed by another ment of end-stage knee OA goes beyond the scope of the task recent meta-analysis in which, on the other hand, the authors make force's commitment. However, there is little doubt that total joint negative conclusions based on selective, and questionable, evidence replacement is cost-effective when all previous modalities have and that this task force does not endorse. Indeed, intra-articular failed and there is significant loss in quality of life, although this is hyaluronic acid injections are relatively safe, although pseudoseptic based mainly on studies other than randomized controlled trials reactions have been reported especially with cross-linked formula- for obvious reasons. Total joint replacement is very effective in tions of the highest molecular weight. In addition, hyaluronic acid relieving severe symptoms of knee OA and has a high benefit/risk induces longer-lasting pain control compared with intra-articular ratio when patients are carefully selected, well informed, anes- corticosteroids and may delay total joint replacement In thesia and surgery are well performed, and rehabilitation is addition, recent findings suggest that there are no significant differ- appropriate . The latter is particularly important when started ences in symptom efficacy compared with oral NSAIDs ; indeed, immediately after surgery and normal activities can be then intra-articular hyaluronic acid might be a good alternative to NSAIDs resumed 6–12 weeks after the intervention, with 95% of all joint in knee OA in older patients or in those at greater risk for NSAID- replacement prostheses expected to survive into the second induced adverse effects.
decade after surgery; only around 20% of patients do not benefit Intra-articular corticosteroids may be suggested in patients following total joint replacement especially when there is an effusion, although this recommenda- Unicompartmental knee replacement is effective when the tion is more theoretical than based on actual evidence. Practically, disease is restricted to a single knee joint compartment and joint aspiration of the synovial fluid is directly followed by patients may experience fewer complications such as deep vein intra-articular administration of corticosteroids, e.g., methylpred- thrombosis . Among other surgical procedures, there is some nisolone acetate or triamcinolone hexacetonide. They indeed evidence for certain efficacy of high tibial osteotomy for medial have higher efficacy than intra-articular hyaluronic acid over compartment knee disease and no favorable evidence for joint the first weeks after administration but their effect on lavage/debridement pain may actually last for only a few (1–3) weeks in controlled For severely symptomatic patients in whom surgery is contra- trials Safety is good, with some evidence also over repeated indicated or if they are unwilling to undergo surgery, the last cycles if scheduled because of the short-lasting analgesic pharmacological chance may be represented by classical oral or transdermal opioids, although their small to moderate efficacy isoutweighed by the large increase in the risk of adverse events Step 3: Last pharmacological attempts before surgery The guidelines for the use of opioid analgesics in the A patient who has failed all the above sequential approaches management of chronic non-cancer pain should be followed may be considered as a candidate to surgery. Last pharmacologicalchances for the severely symptomatic patient may be representedby the recourse to short-term weak opioids. The efficacy of tramadol in relieving pain and improving function in knee osteo-arthritis is small but significant ; adverse events are signifi- Among the many current recommendations and guidelines cantly increased over placebo and (although they may be mostly issued by respectful scientific societies worldwide for the manage- minor) they may lead to treatment withdrawal in a substantial ment of osteoarthritis , this is the first effort to produce a proportion of patients In general, opioids used in arthritis detailed algorithm that guides the physicians through the different patients are associated with significant morbidity and should be steps of treatment of an individual patient.
used with extreme caution Combination therapy with tra- The only previous attempt to generate an algorithm was madol and paracetamol has been shown to be effective when performed with the NICE guidelines . However, this is limited added to NSAIDs in patients who have not adequately responded by regulatory approvals in UK (not all drugs taken into consid- to the latter . However, the safety issues seem to be the same eration here have the same regulatory history in UK than in other for tramadol alone and it is unclear whether combining only European countries) and, especially, by considerations on sustain- paracetamol with NSAIDs in these patients would also result in ability by the UK National Health System (NHS). Indeed, there are a improved efficacy but less adverse effects .
limited number of pharmacological options considered in the final O. Bruyère et al. / Seminars in Arthritis and Rheumatism ] (2014) ]]]–]]] NICE documents. Most importantly, there are only two rings (given should be established beyond the core set and added sequentially the particular shape of the NICE algorithm) after the core inter- at any time throughout prosecution of the algorithm flow, accord- ventions (education, exercise, and weight loss) with low, if any, ing to patient's response.
degree of prioritization in the third and most comprehensive step Step 2 consists of the advanced pharmacological management in that spans from oral NSAIDs to joint arthroplasty. These limitations the persistently symptomatic patient. It is acknowledged here that did not improve with the new NICE guideline update .
the central role within this step is represented by the recourse to oral Other recently released recommendations, such as the Ameri- NSAIDs. Despite the unclear differentiation in the available direct can College of Rheumatology guidelines as it may be the case comparisons between oral NSAIDs and SYSADOAs or topical NSAIDs, in other countries, are also influenced by the incomplete local the task force believes that the former may be more effective in more availability of all drugs considered in our present effort. In severe patients. On the other hand, the most important safety addition, the ACR panel deliberately decided not to develop an considerations were reviewed here and brought to the recommen- algorithm but to use a formal process to rate the available dation that oral NSAIDs should not be used chronically, but inter- evidence. Such process led to the absence of what the process mittently or continuously in longer treatment cycles. Secondly, such designated as "strong recommendation" for any pharmacological safety considerations led to different selection principles among non- interventions for knee OA, except for oral NSAIDs in patients with selective or COX-2-selective NSAIDS, depending on the underlying an inadequate response to acetaminophen (paracetamol) and do gastrointestinal, cardiovascular, or renal risk.
not have contraindications to these agents. Almost all pharmaco- Intra-articular treatment represents a more advanced pharma- logical treatments were rated with a "conditional (weak) recom- cological measure. Indeed, both hyaluronic acid and corticoste- mendation" and some received "no recommendation." The ACR roids have mainly been studied in patients who have failed prior guidelines' authors acknowledged that while most of the recom- analgesic or anti-inflammatory treatment. Both the treatments can mendations will not be controversial, some may be met with be clinically differentiated by the duration of the induced benefit, which is stronger for intra-articular corticosteroids but short-lived Similar limitations are applicable to the new OARSI guidelines.
(a few weeks following a single injection) compared to hyaluro- OARSI has a strong tradition of detailed evidence-based analysis nate whose effect size is smaller (and more controversial) but with clear recommendations on the use of a given intervention – lasting for up to 6 months after a 1–3 weekly treatment course.
In the latest guidelines OARSI has chosen a "minimalist" Step 3 consists of the last pharmacological attempt before approach in which the evidence is only briefly, and often incom- surgery and includes oral weak opioids or duloxetine, whose still pletely, substantiated, and interventions are put in a mere alpha- partial evidence was obtained in non-responders to previous betical order, with the majority of them being classified with an treatment; the mild efficacy is challenged by adverse reactions "uncertain" recommendation. Only a small proportion of them that may lead to frequent patient withdrawal and significant received an "appropriate" recommendation, and knee OA patients with co-morbidities (the vast majority within this elderly patient Finally, end-stage disease management is the last or fourth step population) are left only with topical NSAIDs and intra-articular within the algorithm and consists of surgical procedures mainly corticosteroids as recommended appropriate treatments, beside represented by total joint replacement, with classical opioids as a the possible use of biomechanical interventions and a walking difficult-to-manage alternative for patients in whom surgery is cane . Therefore, the ESCEO task force found not particularly compelling the differentiation between an "appropriate" and an Importantly, the algorithm developed here is to our knowledge "uncertain" recommendation, the latter being difficult to interpret the first attempt to provide guidance on combination or multi- for the prescribing physician.
modal therapy. In fact, first of all, the task force believes that the The algorithm proposed here summarizes the evidence on all core set given here should always be enforced in all patients, proposed treatments and puts them in the sequence suggested by similar to the background treatment foreseen in Step 1, which the evidence itself. Although one major limitation consists in the already provides advice for multimodal therapy by suggesting on lack, in the vast majority of cases, of appropriate clinical studies on one side addition at any time of non-pharmacological treatment treatment prioritization, it is the firm opinion of the task force that modalities when needed and, on the other side, the addition of the evidence from available standard trials is sufficient to propose topical NSAIDs to chronic SYSADOAs (with paracetamol for rescue a given intervention at a certain step of the algorithm.
analgesia) for further pain control if appropriate. When patients Following the core set that, in agreement with most existing are moved to Step 2 and thus to treatment with oral NSAIDs, the recommendations and guidelines, includes information/education, background treatment of Step 1 should not be withdrawn since it weight loss if the patient is overweight, and the establishment of can still provide a different approach to disease and limit the use an appropriate exercise program, the algorithm proposes four of NSAIDs. Similarly, the decision to start intra-articular hyaluro- multimodal steps. During Step 1, there is an attempt to better nate or corticosteroids if the patient is still symptomatic should characterize the use of paracetamol as the initial pharmacological not lead automatically to abandon any of the previous treatment intervention. Given its limited efficacy and some recent concerns modalities. Things get more complicated when the patient enters on its safety, the task force proposes that paracetamol may remain Step 3 because of insufficient symptom control; it may be safe to for rescue analgesia during a background treatment with SYSA- say that while Step 1 modalities might continue because they may DOAs, such as glucosamine sulfate or chondroitin sulfate: both of provide different long-term benefit, the extent to which treat- the latter should be not only of pharmaceutical grade, but specific ments adopted in Step 2 should be replaced depends on the overall prescription drugs are approved in Europe and elsewhere and they benefit/risk ratio. Obviously, Step 4 changes all the perspective, are the only products the task force recommends on the basis of especially when it involves definitive surgical procedures.
the strong clinical evidence, including safety and the possibility of The present effort is an expert consensus strongly based on the additional benefit when used in the early steps of disease manage- available evidence. Among scientific societies within the Rheuma- tology area, ESCEO is the only one devoted to two specific Topical NSAIDs may be added for additional analgesia given disorders, such as osteoporosis and osteoarthritis, given the many their short-term symptomatic efficacy similar to that of their oral biological connections between the bone and the joint. While most counterparts and good systemic safety. Moreover, during this first experts in the present task force have actually clinical and research step, appropriate background non-pharmacological treatment expertise in both diseases or predominantly in OA and have widely O. Bruyère et al. / Seminars in Arthritis and Rheumatism ] (2014) ]]]–]]] published in this field, a small minority have a predominant research interest in osteoporosis. However, all of them are physi- cians specialized within rheumatological areas, have a huge expertise in the care of patients with OA, and sit on scientific boards devoted to OA and pharmacoeconomics of rheumatic When looking at all treatment modalities considered, the task force believes that a complete overview has been provided for non-pharmacological background treatment while, as mentioned, a more extensive review of surgical procedures in Step 4 would have gone beyond the scopes of these recommendations. Con- versely, virtually all pharmacological treatments available have been reviewed and included in the algorithm, with the relevant comments. Drug classes that have not been reviewed here include bisphosphonates, due to failure of the phase III clinical trials with risedronate in controlling symptoms and joint structure disease progression, despite favorable changes in some biomarkers Muscle relaxants have not been included as well, due to the lack of appropriate trials in osteoarthritis, while studies have concen- trated in other rheumatological conditions (e.g., cyclobenzaprine in fibromyalgia).
As any guidance document the present effort suffers from the limitations imposed by the generation of new evidence that may accumulate in the meantime and will need periodic revisions. For the time being, it contains easy-to-follow and evidence-based advice on how and when establishing a treatment flow in patients with knee OA. In addition, this document represents the attempt for a new framework for the development of new international guidelines for the management of OA that is more easily accessible and understandable to specialists and practicing physicians.
[10] National Clinical Guideline Centre. Osteoarthritis: The care and management of osteoarthritis in adults. Clinical guideline CG177; 2014.
O. Bruyère et al. / Seminars in Arthritis and Rheumatism ] (2014) ]]]–]]] [42] Groves T. Effects of glucosamine, chondroitin, or placebo in patients with osteoarthritis of hip or knee: network meta-analysis. Report from post publication review meeting. Br Med J 2011 Avilable from: (Published online: [66] Witter J. Celebrex Capsules (Celecoxib) NDA 20-998/S-009 Medical Officer Review. Available from: 〈 ; 2000 [cited 31.03.2011].
[72] The European Medicines Agency. Assessment report for non-steroidal anti- inflammatory drugs (NSAIDs) and cardiovascular risk. EMA/696137/2012; [54] The European Medicines Agency. Meeting highlights from the Pharmacovi- gilance Risk Assessment Committee (PRAC). Avilable from: [78] US Food and Drug Administration. Information for healthcare professionals: Concomitant use of ibuprofen and aspirin. Available from: 〈 [56] The European Medicines Agency. European Medicines Agency recommends that Protelos/Osseor remain available, but with further restrictions.
Avilable from: 〈 O. Bruyère et al. / Seminars in Arthritis and Rheumatism ] (2014) ]]]–]]]

Source: http://www.md-digital.it/home/videotutorial/03/pdf/Semin.%20Arthritis%20Rheumatism%202014.pdf