Knee OA: does PT help?

 A recent interesting article suggested that physical therapy for knee osteoarthritis (OA) was no better than placebo (see knee arthritis PT not better than placebo AnnRheumDis2022 in dropbox, or 10.1136/annrheumdis2021-221129)

 

Details:

-- 206 Danish adults at least 50 years old in an open label, RCT who had symptomatic and radiologically-confirmed knee osteoarthritis in Denmark were randomized to an eight week exercise and education program versus four intra-articular saline injections, from July 2019 to September 2020

-- average knee pain was at least 4 on a 10-point scale, radiologically verified to be tibiofemoral OA (Kellgren-Lawrence grade at least 2)

-- those with prior intra-articular treatments within three months were excluded

-- participants receiving saline injections were so informed, and all patients were told that the “active ingredients” of both interventions were unverified

-- those in the education/exercise program had two 1.5-hour educational sessions addressing knowledge of knee OA, treatment options, and advice about self-management; the exercise component involved 12 one-hour, group-based, individually supervised sessions, two times a week for six weeks

-- the normal saline injections involved 5 mL injected on weeks one, three, five, and seven using ultrasound imaging guidance

-- all patients were allowed to use mild analgesics (acetaminophen, NSAIDs, aspirin)

-- online questionnaires were emailed weekly from baseline to week eight

-- primary outcome: change from baseline to week nine in the Knee Injury and Osteoarthritis Outcome Score (KOOS) questionnaire pain subscale: 0 (worst)-100 (best)

    -- a difference of 8  KOOS pain points was considered to be significant

-- secondary outcomes: KOOS function and quality of life subscales, and patients global assessment of disease impact

 

Results:

-- treatment adherence the two groups with similar

-- primary outcome, KOOS pain scale: least-squares mean changes were 10.0 ± 1.5  for exercise and education and 7.3 ± 1.5 for saline injections

    -- difference of 2.7 points (-0.6 to 6.0), p=0.11 for test of superiority

    -- test for equivalency: p=0.0008

 

 

-- all group differences in key secondary outcomes were within the predefined equivalence margins at week 9

    -- KOOS function score: 0.8 points (-2.3 to 4.0)

    -- KOOS quality-of-life score: 1.8 points (-1.5 to 5.2)

    -- Participant Global Assessment: 5.7 mm (-11.3 to -0.1)

-- adverse events and serious adverse events were similar between the groups

-- the results of both the primary and key secondary outcomes were unchanged at week 12 as well

  

Commentary:

-- the basic concept of the study was that one could not compare a PT intervention versus nothing (too much placebo effect), so the next best thing was to do an inert/placebo injection (normal saline)

-- those with knee OA often do have fluctuating courses. in this study, it is likely that knee OA symptoms were worse at the time of recruitment (hence their interest in the study) but may well have improved spontaneously over time without any therapy: ie there was a selection bias towards recruiting people who have more severe symptomatic knee pain, which as a group will have a tendency to get better over time given the individual pain variability (ie regression to the mean). And this would apply to all interventions, including PT or injections

-- there are more than 100 studies in the literature finding that exercise was beneficial for knee OA, and physical therapy is one of the bedrocks of standard knee OA treatment. This study clearly questions that result, perhaps reflecting the fact that physical therapy is prescribed when patients have a lot of pain, and improvement may simply reflect this regression to the mean

-- that all being said, if the placebo effect plays an important role, physical therapy may in fact be better than doing nothing. It might be useful to have studies looking at alternative active treatments, including cognitive behavioral therapy, mindfulness, home-based exercises focusing on quad strengthening, or medications (NSAIDs, acetaminophen, chondroitin sulfate/glucosamine, etc)

-- we do all know that the placebo effect can be quite profound:

    -- there are an array of prior blogs on placebos: see http://gmodestmedblogs.blogspot.com/search?q=placebo

    -- in particular there are interesting ones on the genetics of the placebo effect, related to specific genes in the “placebome”: see http://gmodestmedblogs.blogspot.com/2015/04/placebo-genetics-and-placebome.html

    -- another study a few years ago found that patients with chronic low back pain, where both they and their clinicians were aware that the intervention was a placebo, still helped a lot (and the patients stated that they knew that the intervention “was not an active substance”). And, many participants requested continuing the placebo after the end of the study: see http://gmodestmedblogs.blogspot.com/2021/06/placebos-my-favorite-blog.html

-- this current study defined the minimal clinically important difference in the KOOS scales as 8 points. Some of the studies have used a 10-point margin. so, this study was m ore likely to find a differential effect at this lower limit (but did not)

-- a prior study, the Strength Training for Arthritis Trial (START), had a more long-term assessment, finding that 18 months of muscle strengthening exercise for patients with knee OA, including both high- and low-intensity strength training, was not more effective than attention control in 377 participants; the attention control group attended 60-minute group workshops biweekly for the first six months and monthly thereafter (see Effect of High-Intensity Strength Training on Knee Pain and Knee Joint Compressive Forces Among Adults With Knee Osteoarthritis: The START Randomized Clinical Trial | Geriatrics | JAMA | JAMA Network )

-- an interesting systematic review and meta-analysis of 38 RCTs did find that intra-articular saline for knee OA had long-term benefits in pain relief for 6 to 12 months (see Clinical benefit of intra-articular saline as a comparator in clinical trials of knee osteoarthritis treatments: A systematic review and meta-analysis of randomized trials - ScienceDirect ). Perhaps the saline injections have physiologic mechanisms besides being "inert" placebos??

-- and, one other issue concerns the type of PT done. there are some studies suggesting that individualized PT programs work better than group ones, as found in  https://www.oarsijournal.com/article/S1063-4584(19)31087-8/fulltext. the study above had group sessions with individual supervisions, sort of a combo PT

 

Limitations:

-- this is a short-term study, though the START trial did find that 18 months of muscle strengthening exercises for patients with knee OA was not more effective than an attention control

    -- and, as noted above, intra-articular saline injection for knee osteoarthritis may last for 6 to 12 months

-- this was an open-label study, where participants knew what they were getting, fundamentally different from a double-blind randomized controlled trial. However, this open-label study design allowed a more real-world assessment of interventions. And patients involved in the study all likely felt they were getting an intervention, versus the uncertainty (and perhaps second-guessing) of whether they were assigned a placebo arm in a traditional RCT, though in this current study they were all told that the different arms of the trial were "unverified

-- though many of the results were subjective (pain scale, etc), there were several that were semiobjective, such as the fast walk test, chair stand test, and climbing test; all of these found no difference between the interventions

-- the placebo effect is an important one in all trials as well as in clinical practice. However there may be different weights given to different interventions. For example, in this study those assigned to education/exercise had lots of attention from the caregivers, known to improve the placebo effect. On the other hand, those having an invasive intervention such as knee injection might have a different degree of placebo effect, which might in fact be more profound for some patients

    -- there was an interesting article on knee OA assessing different placebos, finding for example that intra-articular placebo had a statistically significant effect size of 0.58 vs 0.20 for oral placebo (and topical placebo was also better than oral placebo): see placebo treatment differences AIM2015 in dropbox, or doi:10.7326/M15-0623

    -- and perhaps having had 4 injections within 1 week as in the above study reinforced the placebo effect even more

-- also as mentioned above, perhaps the results would have been different with an individualized PT program as opposed to a group-based one

so, an interesting study questioning the utility of PT in those with knee OA. a few comments:

-- one reason i bring this study up is that it questions long-held assumptions, which really do need to be formally tested

-- our understanding of the complexities of the human body are often lacking: those of us in primary care often see people with lots of pain that simply gets better with reassurance and watchful waiting. for example, i have had 2 patients with disabling lumbar spinal stenosis with severe functional limitations, where the pain simply vanished over time. i have no idea why, but i know that if they had gone to interventional radiology for steroid injections and got better, that would have reinforced my (and their) sense that the injections really worked. in fact, i discussed the potential "spontaneous" recovery in my first patient with the second patient, who elected to just take pain pills, but then weaned herself off them completely.

-- and, another but related issue in terms of our (perhaps false) assumptions: it turns out that there really is not such a great relationship between clinical symptoms of knee OA and radiologic findings, as found in several studies: see https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2542996/

-- there are other studies in the literature about other orthopedic pain (eg shoulder pain), where PT did not help much

-- so, how should this study change our practice??? Not sure. some patients probably do as well with a home-based exercise program (though many may need the structure of PT or classes). Many patients (at least that i take care of) who go to PT for several weeks do not continue the exercises at home after the PT ends (and many do fine). 

-- But my broader sense is that it is really important for pretty much all patients to stress the importance of exercise overall. and perhaps the best outcome in those with knee OA or other orthopedic issues is integrating some specific stretching/strengthening exercises into a regular exercise program (ie, advocating a regular exercise/walking schedule with all patients but including some specific stretching as part, instead of just doing stretching/strengthening because of back pain or knee pain, where the subliminal message is that it is not needed if the pain is not too bad)

geoff

 

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@uphams.org

 

to get access to all of the blogs (2 options):

1. go to http://gmodestmedblogs.blogspot.com/ to see them in reverse chronological order

2. click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category​

3. or you can just click on the magnifying glass on top right, then  type in a name in the search box and get all the blogs with that name in them

 

or: go to https://www.bucommunitymedicine.org/ , a website from the Community Medicine section at Boston Medical Center.  This site does have a very searchable and accessible list of my blogs (though there have been a few that did not upload over the last year or two). but overall it is much easier to view blogs and displays more at a time.

 

 

please feel free to circulate this to others. also, if you send me their emails, i can add them to the list