intra-articular steroids??

another article came out expressing concern that intra-articular steroids may cause accelerated joint destruction (see knee arthritis steroids not safe radiol2019 in dropbox, or doi/10.1148/radiol.2019190341).  this one made it onto NPR, with a significant warning about doing these injections.   This article was not a study, but a note of caution about using intra-articular corticosteroid (IACS) injections in the hip and knee, as follows:

--they note that several academic organizations (Am Coll of Rheumatology, Osteoarthritis Research Society International) have conditional support for IACS, some (Am Acad of Orthopedics) do not have recommendations (inconclusive evidence, in their 2013 guidelines: https://www.aaos.org/cc_files/aaosorg/research/guidelines/treatmentofosteoarthritisofthekneeguideline.pdf)

--a Cochrane review in 2015 of 27 trials (1767 patients) found low quality of evidence, with the conclusion, per the current authors, that IACS injections "might have resulted in a moderate improvement in pain and a small improvement in physical function; however the quality of the evidence was low, and the overall results were inconclusive", and that IACS injections "appeared to cause as many side effects as the placebo (13% vs 15%)"

--they do make it clear that patient preference "should have substantial influence on the type of treatment selected" and many patients are "not suitable candidates for joint replacement because of their older age, comorbidities, or both"

--they also reinforce that anticoagulation treatment is not a general contraindication for IACS injections, though unstable coagulopathy is

--lab studies with in vitro chondrocytes do find an adverse effect on cartilage by steroids, esp at higher doses

--there are radiologic studies finding increased osteoarthritis (OA) in those getting injections [though, see comments below. and is there more OA because people are able to weight bear more and do more active activity after an injection?]

--they do have some anecdotal observations, using triamcinolone 40mg plus lidocaine 1% 2ml plus bupivicaine 0.25% 2ml [lots of local anesthetic!!], the adverse outcomes of: accelerated OA progression (6%), subchoncral insufficiency fracture (0.9%), complications of osteonecrosis (0.7%), and rapid joint destruction (0.7%) [again, not a randomized study, with very likely selection bias in terms of referrals to a tertiary care center, attribution bias, etc], with the following comments:

    --accelerated OA progression: no clear definition of what this is (some state empirically that it is >2mm within 12 months, though no agreed-upon definition). assessment can be difficult, since minor changes in patient positioning can result in major observed changes in joint thickness. and, accelerated OA progression occurs in those without IACS and may be related to subchondral fractures. happens more frequently in elderly women or those who have sustained trauma. so, at least to my reading, not clearly associated with IACS (see https://www.ncbi.nlm.nih.gov/pubmed/28689367 , which states in the full article "the role of intraarticular steroid and anesthetic injection as a potential cause of the development of RPOA (rapidly progressive OA) has not been systematically evaluated. any relationship between IACS injection and the onset is likely temporal only"

    --subchondral insufficiency fracture: these are fractures typically of weakened bone (eg osteoporosis, or demineralization from disuse), though have been found in active younger adults. patients present with acute pain typically in weight-bearing joints, gets worse for weeks, and without intentifiable trauma. can progress to articular surface collapse (though early on, can have normal xrays), and can lead to spontaneous octeonecrosis of the knee. but also can heal spontaneously. MRI can diagnose. 

    --complications of osteonecrosis. esp the femoral head and condyles. patient typically has pain. often radiographically occult and needs MRI for diagnosis. when there is articular surface collapse, secondary OA and persistent pain can develop, necessitating joint replacement. no known contraindication or benefit of IACS

    --rapid joint destruction: can happen with IACS, put also without prior intervention or underlying disease

--and, they do note that "there are no large (>200 subjects) retrospecitve reviews or randomized controlled studies with long-term (>1 yr) follow-up"

Commentary:

--i bring up this article partly because it made the news (which cast IACS in a negative light) and largely because i have seen such dramatic functional relief for the patients i have injected

--see http://gmodestmedblogs.blogspot.com/2017/05/steroid-knee-injections-do-they-help.html , which critiques one of the better, further elaborated in this blog:

    --this study found that patients with symptomatic knee OA and synovitis on ultrasound had more cartilage loss with triamcinolone 40mg vs saline every 3 months for 2 years and MRI to assess articular structures, and no significant difference in pain scores, but:

        --the patients had only mild symptoms before the study

        --the cartilage loss is hard to assess: there is no accepted minimal clinically important differences by MRI assessment

        --pain relief was assessed at each 12 week visit, without data on how patients did clinically prior to the 3-month mark

           --in my pretty large injection practice, with 1000-2000 knee injections over many years, a significant number do have waning effectiveness in the 2-3 month post-injection period. so not so surprising that there was no statistically signficant difference when checked at 3 months later

        --are there confounding issues: do those getting triamcinolone actually get dramatic relief in the first 2 months (as the vast majority of my patients do), then exercise much more (a good thing) and this exercise and the associated knee trauma causes the cartilage destruction??? or they noted decrease in A1c (0.3 percentage points) in the triamcinolone group: does this play a role or reflect other unaddressed potential confounders??

    --one relevant side-issue is potential confounding of the effects of steroids vs local anesthetics (as typically given in IACS) or with saline (as in the above study):

        --there are some studies finding that the "placebo" injection of saline does seem to have some benefit on pain symptoms. maybe a better study, at least for pain, would be: injected triamcinolone vs sham injection.  or even injected saline vs sham injection

        -- there are several studies suggesting that local anesthetics themselves may lead to significant cartilage loss: see https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4796530/ for a review of the effect of local anesthetics (lidocaine, bupivacaine etc) on cartilage, suggesting significant chondrolysis. there is some suggestion that bupivacaine 0.13% or less causes less cartilage damage. Perhaps this is a safer local anesthetic to use than the lidocaine 1% that i have been using, though my initial investigation is that it is very expensive and hard to get in outpatient clinics (I will pursue this more…)

--the authors of the current article do note that the pain benefit from intra-articular steroids may still be important for those patients unable or reluctant to have surgery.  i would second (third, and fourth) this comment: there is no question in my mind that exercise and functional ability to walk are extremely important quality (and quantity) of life issues. and many many patients do fantastically well with steroid injections (though the effect after many injections may wane). i sometimes even inject steroids more often that every 3 months in order to preserve walking abillity in those not candidates for surgery (though i do suggest more strongly the advantages of surgery in those with waning benefits from steroids).  

    --for a person doing primary care, there are relatively few interventions we do that causes the patient to smile and jump up in jubilation....  and steroid injections is one of them

--and, of their anecdotal observations noted in the above study (accelerated OA progression, subchondral insufficiency fracture, complications of osteonecrosis, and rapid joint destruction), it is not clear to me that any of them are causally linked to IACS. Though i personally do not do intra-articular hip injections, i do think that with the frailty of arterial supply to the hip joint and potential for developing avascular necrosis that these patients need ultrasound-guided injections.  though i do inject surrounding structures (trochanteric bursitis, superficial trigger points)

-- In terms of getting xrays to check the extent of OA of the knee (or even the relevance of xrays clinically), a review of knee OA by a rheumatologist involved in the prospective Framingham Study and research on knee OA notes that "radiographic findings correlate poorly with the severity of pain and radiographs may be normal in  persons with disease". in a personal conversation with him, he felt there was no added value to xrays, unless the clinical history is less suggestive of OA (eg knee pain persists after effective OA therapy, the knee pain is nocturnal or not activity related). see knee arthritis felson nejm 2006 in dropbox, or Felson DT. N Engl J Med 2006; 354: 841. so, i have not been doing routine xrays. And there are no recommendations to do xrays prior to IACS injections. These authors above argue that in those with subchondral insufficiency fractures, steroids might inhibit healing, or might decrease pain enough that there is more joint loading/weight bearing and more likely joint collapse. the treatment should be more just supportive care and possibly bisphosphonates or prostacycin analogs. but i have not found any clear documentation of this risk; and, xrays may be normal early in the fracture; perhaps there is some benefit in doing repeated xrays prior to subsequent injections to see if there is accelarated joint destruction. but, again, this association is largely hypothetical and devoid of clear data.  and once there is osteonecrosis, steroids may in fact relieve pain and help the patient, though the ultimate relief is likely to be joint replacement

-- a few caveats about IACS that are well-documented:

    -- IACS should not be given for at least 3 months prior to a joint replacement because of increased risk of infection

    -- one should not do IACS in HIV patients on ritonavir or cobicistat, given risk of severe cushings after injections (see http://gmodestmedblogs.blogspot.com/2019/05/hiv-meds-local-steroids-and-cushings.html )

So, I do think that IACS is an important and highly clinically effective treatment for OA, as I have found having done lots of knee injections, typically giving patients back their normal functional potential for walking and exercise in general, and with the attendant benefits on quality and quantity of life.  Sometimes an injection lasts a really long time (many >1yr, occasionally just a 1-time event), more often in the 3+ month range. I do think it is important to couple injections with PT, or at least with exercises to improve the supporting knee muscles (esp quadriceps strengthening exercises). So I am hesitant to accept an article as above, largely based on anecdotal findings and inadequate data, and am afraid such articles in the popular press might dissuade patients from getting these helpful injections. Though I do think it is important that patients understand that the long-term risks of IACS are not clear, especially in those receiving multiple injections…

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:

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

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