Talk

Advanced search

Anyone any good at evaluating summaries of scientific studies?

(7 Posts)
LaurieFairyCake Wed 31-Jul-19 08:15:04

6 weeks ago I had steroid injections in my knee - been pain free (was in agony).

It's starting to come back. Physio suggested hyaluronic acid injections.

There are multiple blurbs from different clinics all over the country saying they work, '30% of patients become pain free for 2 years' etc etc hyperbole

However they are not NICE recommended and this article on actual scientific studies is what I want someone else's opinion on

https://www.google.com/amp/s/www.forbes.com/sites/stevensalzberg/2015/05/18/do-injections-work-for-knee-pain-dont-waste-your-money/amp/

I can only have steroid injections every 3 months so am doing everything recommended to hold the pain at bay (at huge financial cost)

I don't want to have injections that don't work and I think that article is pretty bloody clear.

OP’s posts: |
MedSchoolRat Wed 31-Jul-19 16:45:22

Me... I do this for a living. Give me a sec...

CMOTDibbler Wed 31-Jul-19 16:53:44

A quick look at the published data on Pubmed, including a trial where they actually compared placebo injection vs the hyaluronic acid injection seems to show that they have no effect beyond the placebo (which of course shouldn't be ignored as pain relief is great, but injections into joints aren't without risk)

MedSchoolRat Wed 31-Jul-19 16:55:33

Short answer: yup, ineffective.
Happy to provide more detail why I agree about it being ineffective. Disappointing, I'm sure.
The full article is here.

SirTobyBelch Wed 31-Jul-19 17:44:38

As it happens, I was talking to a rheumatologist about this earlier today. She said they aren't viewed as being effective and only continue to be used by a few orthopaedic surgeons for private patients.

There is a divergence of views in the medical research literature. See, for example Maheu et al 2019 and Altman et al 2018, which suggest it is effective. However, most reviews of the research indicate that it is more effective in particular groups and less (or not) effective in other groups, particularly older patients with more severe disease. Many of the studies have flaws in their design or analysis, so the picture isn't really clear yet. Clearly, hyaluronic acid has fewer side-effects than hydrocortisone but that's only worth taking into consideration if it's going to have some beneficial effect. I'm afraid there isn't a simple answer to your question.

MedSchoolRat Wed 31-Jul-19 18:30:58

Ok, I'm going to be a snob about this.
OP linked to an blog reviewing article by Rutjes et al.

Rutjes 2012 is a systematic review of many outcomes (not just pain). Systematic review is supposed to mean they made big effort to gather every bit of evidence there was. They published in a very top notch journal. What that means is the journal stakes a lot of their reputation on getting the best evidence, they will tend to get the best peer reviewers who hauled the authors over the coals. Rutjes described results from 89 RCTs & 12 667 patients. 71 of these trials addressed intensity of pain (9617 patients). RCTs are the best quality experimental evidence. Rutjes pooled the original numbers from the original studies to find a net effect on all patients. They identify the best quality studies and pool only their results. They went looking for missing studies ("publication bias" against negative results). Funded by the Arco Foundation.

Altman et al 2018 is also a systematic review, in a medium impact journal (SAR). Altman et al only looked at pain outcomes. They found only 10 RCTs and 7 other study designs = "observational" studies which can have more biases than RCTs. They claimed that a quality of evidence assessment ("risk of bias") was impossible because of the diversity of the studies they found (this claim is very untrue). They talk about the different studies only with words (no pooling the original numbers). Their study was funded by Ferring Pharmaceuticals, Inc.

Maheu 2019 only has the advantage of being very recent. They
published in same journal as Altman 2018. This is a meta-review= a review of the reviews. There is NO QUALITY grading for how good any one of these reviews was (AMSTAR is a quality checklist they could have used). All or most authors have long list of conflicts of interest declared (Big Pharma).

The editor in Chief of SAR has many financial links to Big Pharma.

There are ways to tell which study you should believe.
Rutjes is by far the better quality study over the other articles.

LiliesAndChocolate Wed 31-Jul-19 20:40:48

There is a more recent articles from the American Academy of Orthopaedic Surgeons
Viscosupplementation for Osteoarthritis of the Knee
A Systematic Review of the Evidence 2015
www.ncbi.nlm.nih.gov/pubmed/26677239

It confirms previous meta analysis.

number of systematic reviews addressing intra-articular administration of hyaluronic acid (HA) have been performed and reported6-10. Those reviews have shown significance with respect to symptom relief, but not all have taken into account themagnitude of the effect. Three systematic reviews utilizing clinical significance were unable to demonstrate efficacy with respect to pain relief 6,7,10. The authors of all reviews commented on the substantial publication bias and heterogeneity of the clinical trials of intra-articular HA. In an attempt to minimize heterogeneity, we assessed the effect of intra-articular HA for knee osteoarthritis using a “best-evidence” systematic review as described by Slavin11. .../......./.... In conclusion, this best-evidence systematic review as- sessing the clinical significance of outcomes involving pain relief and functional improvement does not support the rou- tine use of intra-articular HA. In contrast to previous reviews, we found no significant evidence of publication bias in the studies that we selected for analysis. The patient benefit of intra-articular HA was not clinically important when compared with intra-articular saline solution injections used as a placebo. Subdividing HA preparations by molecular weight did not change the results of the analyses. Selecting the best evidence resulted in significantly reduced heterogeneity but did not change the outcome; no clinically important improvement in pain and other outcomes from a patient’s perspective was found.

Join the discussion

Registering is free, quick, and means you can join in the discussion, watch threads, get discounts, win prizes and lots more.

Get started »