Tag Archives: Cochrane

General

Straight Talk Tuesday; Caroline Fife

Comment

Carolyne Fife demonstrates in her columns “the horror of providing wound care”. In two blogs (below) she describes the impact the apparent low stature of wound healing has. Basically; the lack of underpinning of interventions leads to a dramatic reduction of funding. As it should. However, in the case of wound care the lack of underpinning is not a result of lack of efficacy but a lack of research outcomes. So here the reduction of funding is a result of the dysfunction of research. The cause is very simple; having a wound is generally regarded a complication and therefore not interesting. Sadly society and (have to say) most doctors are not interested, lack knowledge, do no research and are astoundingly unaware of the numbers. As a result, the science of wound healing is under-represented and under-developed in the medical profession. (Why don’t we have regeneration specialists? ) Therefore, even if Caroline would have generated data and healed patients (which should not be the same) we still lack a proper framework to translate findings in a wider context. This would be no problem if having a wound was a rare phenomenon. But it is not, wounds lead to massive human suffering and consume a relatively big part of health care spending. But the latter is a financial issue. Luckily the financial guru’s know that if you cut the budget the cost will go down, as simple as that. Doctors agree and push the patients out of their budget (back to home care) asap. The fact that budget cuts cause massive suffering (but numbers do not cry or die) and that the costs resurface somewhere else in the system in a 10-fold time or so is not interesting because that is NIMB-cost. (Not In My Budget) And since 2008 we all know who pays for the final NIMB budget.

Straight Talk Tuesday

The second blog mentioned is below. Is this what you get when one applies regulations instead of knowledge: if you only have a hammer, every problem looks like a nail? Is this the future of EU healthcare funding?

Monday Musings

Monday Musings

Level 4

The question

Comment

There is one question wound care specialists typically do not ask. If Winter and Hinman proved the use of cling film speeds up wound healing by 40%, why are there hardly any studies repeating the phenomenon in a clinical setting and do meta-analyses invariably deliver hardly any or no effect.  It seems to me the gap between Winter (animal research, evidence level C), Hinman (evidence level B) +40% and meta-analyses  (evidence level A1) +/- 0% requires an explanation. To me it is a logical question for a logical problem: the gap is too large, we cannot find 40% and 0% outcome at the same time. So who sheds some light on the logic behind the phenomenon?

General, Level 4

Epic Fail: Hyperbaric oxygen therapy for treating chronic wounds

Comment

Cochrane publiceerde de review “Hyperbare zuurstoftherapie voor de behandeling van chronische wonden” (hier is het originele artikel)

Dit zijn de resultaten:

  • Voor diabetische ulcera lijkt HBOT de kans op genezing op korte termijn te verbeteren (tot zes weken) maar niet op langere termijn follow-up. HBOT kan het aantal grote amputaties bij mensen met diabetes die chronische voetulcera hebben verminderen.
  • Voor chronische wonden met vasculaire oorzaak kan HBOT kan de wond  verkleinen.
  • Voor chronische arteriële ulcera of chronische decubitus was geen bewijs van de werkzaamheid van HBOT.

Deze resultaten zijn te verwachten. Niet omdat HBOT niet werkt, maar omdat de onderzoekers cruciale informatie missen. Dit doen zij omdat het niet beschikbaar is. De Cochrane review laat op onvermijdelijke en reproduceerbare wijze zien dat er vrijwel geen bewijs te vinden is, maar dat ligt niet aan de Cochrane stichting.

De problemen beginnen met de inclusie criteria. Dus nog voor er ook maar iets is getest. Als het inclusie criterium “een niet genezende wond” is, bent je vanaf het begin verloren. En dat is wat ongeveer alle studies gemeen hebben. (De echte criteria staan onder) Wonden kunnen om heel veel redenen niet genezen. Deze redenen kunnen patiëntgebonden, wond gerelateerd, mechanisch of metabool zijn. (Level 1,2,3 en 4, die meestal onafhankelijk zijn.)

Allereerst moeten we ons realiseren dat bij een chronische wond datgene wat de wond veroorzaakte niet noodzakelijk datgene is wat het genezen tegenhoudt. Toch gooien we allen wonden in een studie op een hoop: niet genezend. Vaak wordt er nog een indeling gemaakt op oorzaak van de wond (logische fout “1”). Probeer dan maar eens aan te tonen dat jouw interventie x is werkt. Dat is een schending van de werkelijkheid.

Als resultaat kunt je, als je geluk hebt/per ongeluk, ontdekken hoeveel van deze niet-genezende wonden reageren op je interventie  passen. Dat is heel wat anders dan het effect van je interventie aantonen. (logische fout “2”) Het is door de opzet van de studie al bijna onmogelijk om te ontdekken hoe effectief de interventie is. Omdat we geen goede gereedschappen hebben om de oorzaak van de wond te ontdekken hebben we helaas ook geen andere keuze dan de wonden maar bij elkaar op een hoop te vegen. Het is misschien wel verontrustend dat we op voorhand ons niet realiseren dat de kans op succes daarmee vrijwel 0 is. (logische fout 1 + 2)

En nu het echte probleem: het zou heel interessant zijn om erachter te komen welke wonden reageren; en waarom en hoe ze dat doen. Dit helpt om meer oorzaken van het openblijven van de wond te vinden. Dan zouden we deze oorzaken kunnen herkennen voor we weer eens een dure behandeling toepassen.

Als je hulp nodig hebt: harmjsmit@gmail.com

Inclusion criteria varied in these trials. Doctor 1992 included any person with diabetes with a chronic foot lesion (time not specified); Faglia 1996a included people with diabetes and Wagner grade 2, 3 or 4 lesions (Wagner 1987); Lin 2001 and Kessler 2003 people with “early diabetic feet”, Wagner grades 0, 1 or 2, while Duzgun 2008; Abidia 2003 and Londahl 2010 included people with diabetes whose lesions had been present for more than four weeks, six weeks and three months respectively. In addition, Londahl 2010
required evidence of good standard wound care in a specialist clinic setting for a minimum of two months. Exclusion criteria generally followed from the specific inclusions detailed above, but Abidia 2003 also specifically excluded participants for whom vascular surgical procedures were planned and Kessler 2003 excluded all patients with transcutaneous oxygen tensions of < 30 mmHg. Ma 2013 included patients with diagnosed diabetes, at least one fullthickness wound below the ankle (Wagner grades III or less) for > 3 month, standard care for > 2 month, TcPO2 > 30 mmHg. Khandelwal 2013 included patients with a diabetic foot ulcer of at least 8 weeks duration, patients with only stage III and IV diabetic foot ulcer and the absence of vascular insufficiency.

General, Level 4

Epic failure: Hyperbaric oxygen therapy for treating chronic wounds

Comment

The Cochrane published the review Hyperbaric oxygen therapy for treating chronic wounds (original article)

Here are the results:

  • For diabetes-related foot ulcers, we found that HBOT seemed to improve the chance of healing in the short term (up to six weeks), but not with longer term follow-up. HBOT may reduce the number of major amputations in people with diabetes who have chronic foot ulcers.
  • For chronic wounds caused by a disease to the veins of the leg, we found that HBOT may reduce the size of wounds.
  • For chronic wounds caused by lack of blood supply through the arteries or chronic pressure ulcers, we found no evidence to confirm or refute any effects of HBOT.

These results are to be expected. Not because HBOT is not working, but because researchers lack critical information, simply because it is not available. The Cochrane results are inevitable and reproducible non-conclusive but Cochrane is not to blame. Let’s explain two of the many logical errors causing the failure of wound care research.

It starts with the inclusion criteria. If your inclusion criterion is “a non-healing wound”, you are lost from the start. (the real list is below) There are many reasons wounds do not heal. They can, for example, be patient related, wound related, mechanical or metabolic. (Levels 1,2,3 and 4 which are usually independent of each other)

The first notion regarding chronic wound is that wat caused it, is not necessarily preventing it from closing. Yet we place them all on a pile, sorted by cause of the wound (logical error 1) and then investigate if intervention x is functional. So you are testing in the blind, which is very different from double-blind. That is a gross neglect of reality.

Therefore, if you are lucky,  you will discover how many non-healing wounds accidentally fit your solution (logical error 2). This is very different from finding how effective your intervention is. In today’s setup, it is therefore almost impossible to discover how effective your intervention is. Because we lack markers and/or knowledge to diagnose and characterize the cause of not healing. Therefore, we have no choice but to pile wounds together. It is a pity we do not recognize this beforehand. If you cannot select the wounds which are lacking the solution you provide, the chances of success are next to 0. (logical error 1+2)

And now the real issue: it would be very interesting to figure out which wounds respond. And why and how they respond. This will enable us to recognise this the next time before we apply an expensive treatment. Or in other words, the non-responders are more interesting than the responders. Sadly research is acting in exact the opposite way.

It starts with the inclusion criteria. If your inclusion criterion is “a non-healing wound”, you are lost from the start. (the real list is below) There are many reasons wounds do not heal. They can be patient related, wound related, mechanical or metabolic. (Levels 1,2,3 and 4 which are usually independent of each other)

 

And now the real issue: it would be very interesting to figure out which wounds respond. And why and how they respond. This will enable us to recognise this the next time before we apply an expensive treatment.

If you need help: harmjsmit@gmail.com

 

Inclusion criteria varied in these trials. Doctor 1992 included any person with diabetes with a chronic foot lesion (time not specified); Faglia 1996a included people with diabetes and Wagner grade 2, 3 or 4 lesions (Wagner 1987); Lin 2001 and Kessler 2003 people with “early diabetic feet”, Wagner grades 0, 1 or 2, while Duzgun 2008; Abidia 2003 and Londahl 2010 included people with diabetes whose lesions had been present for more than four weeks, six weeks and three months respectively. In addition, Londahl 2010
required evidence of good standard wound care in a specialist clinic setting for a minimum of two months. Exclusion criteria generally followed from the specific inclusions detailed above, but Abidia 2003 also specifically excluded participants for whom vascular surgical procedures were planned and Kessler 2003 excluded all patients with transcutaneous oxygen tensions of < 30 mmHg. Ma 2013 included patients with diagnosed diabetes, at least one fullthickness wound below the ankle (Wagner grades III or less) for > 3 month, standard care for > 2 month, TcPO2 > 30 mmHg. Khandelwal 2013 included patients with a diabetic foot ulcer of at least 8 weeks duration, patients with only stage III and IV diabetic foot ulcer and the absence of vascular insufficiency.

General

Low reproducibility rates within life science research undermine cumulative knowledge production

Comment

To cite the article below: “Low reproducibility rates within life science research undermine cumulative knowledge production“. Wound care research provides an excellent example. Quoting the EWMA Study Recommendations for clinical investigations in leg ulcers and wound care (2014) “However, we are all aware that the quality of many studies in this field remains poor and we would be doing a disservice as the European Wound Management Association if we did not encourage our members to join in the challenge of raising the quality of studies for the benefit of our patients.” This is 2015!

Wound care research more than often (def)end inconclusive articles with the phrase : “more research is needed”. The result may be inconclusive but the expert opinion is not, research was flawed because it does not proved the result the author was expecting based upon his or her expert opinion. In this little sentence you may already feel the importance of the expert opinion. It is interesting to notice meta analyses with conclusive results are put aside to the benefit of expert opinions. The expert opinion is leading in wound care. This means the wound care specialist concludes the meta analyses are based on wrong data and prefer their own opinion (expert opinion).

The “expert opinion” leads to the following observation: either it is impossible to do wound care research or we have a massive logical error preventing us to come up with widely accepted meta analyses. Anyway, following the international guidelines meta analyses are leading, not the expert opinion. But no matter who is right, in 2015 we still are using the same paradigms as in 1943.

So the conclusion of this article is valid for wound care; yes, there has been no cumulative knowledge production in wound care. (for 70 years)

Freedman LP, Cockburn IM, Simcoe TS (2015) The Economics of Reproducibility in Preclinical Research. PLoS Biol 13(6): e1002165. doi:10.1371/journal.pbio.1002165
http://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.1002165
http://ewma.org/english/publications/ewma-documents-and-articles/ewma-study-recommendations.html
E L Howes. The rate and nature of epithelization in wounds with loss of substance SGO 1943 Vol 76 (738-745)

below are two random articles, feel free to do your own research…
http://www.woundsresearch.com/article/real-world-experience-decellularized-dehydrated-human-amniotic-membrane-allograft
http://www.magonlinelibrary.com/doi/10.12968/jowc.2015.24.6.245

Level 4

Reverse evidence

Comment

Since 1962 we all know that wounds without a scab heal 40% faster compared to wounds with a scab. This has been tested on two pigs (Winter, Nature 1962) and seven inmates (Hinman, Nature1963). This means, since 1962, we all are convinced that a for instance diabetic wound on an 83-year-old will heal up to 40% faster, only if you apply a moist environment.

Reversibly you may also try to answer the question how much slower wounds would heal if you would use of moist gauze, which have to be changed up to 3 times a day. Almost 50 years after Winter some people have tried to answer this question (Ubbink DT et al, Arch Surg 2008). What appears was that moist wound healing was not better but actually worse than a conventional gauze treatment. But there was one remark, these findings apply only for wounds with an acute etiology (and in a hospital setting).

Ehh well yess…. what type of wounds were investigated in the Winter and Hinman papers?

But fair is fair, one study should not make a difference, scientific principles dictate a finding should be repeated several times in order to be accepted.

This is why we have organisations which examine the evidence. One of those organisation is the Cochrane foundation which has as purpose to help medical professionals in their decisions. The method they use is to ruthlessly examine all research papers against the highest standards. <http://www.cochrane.org/about-us>. And nothing is easier than to go to their website and read what they have to tell about wound care.

If I cite them randomly: negative pressure therapy, there is not enough evidence. (And it is not only Cochraine but also for instance Vig S et al, J. of Tissue Viability 2011), Alginate; there is no evidence that alginate is better than any other addressing, foams; conclusion is the same, no evidence. The only exception appears to be hydrogel were the conclusion is that hydrogel is have some evidence to be more effective compared to other dressings. Silver dressings, no evidence etc. After a quick glance I counted 105 studies with almost 10,000 patients which actually met the Cochrane criteria.

Cochrane provides one, real unnerving outcome, no one is able to prove or repeat the 40% faster promise from the original articles more than 50 years. Apparently it is not possible to design a study or do a study which is meeting current scientific criteria and show a 40% faster healing.

So we start reducing our expectations. Perhaps we are asking too much my looking for a 40% faster healing, so let’s settle for 20%, 10% or even 5%. That appears fair to me.

But even that is not possible, the most positive statement in these studies is “there is some proof”, which is quite different from 5% faster healing. Apparently it is not possible for any current wound care dressing to prove in a well-designed study it actually makes sense to use.

Now what, it appears we have a problem.

We are not able to prove how much a wound benefits from using a moist dressing. The next step would be to reverse the question, does it actually work? The answer to this question is very easy to answer, because there are plenty studies, meeting the highest scientific standards providing the answer, the Cochrane collection from above. The significant answer meeting the highest standards is: a moist environment does not lead to faster wound healing.

And this finally leads us to the heart of the problem. We have a Mexican standoff between 2 pigs, seven inmates and 285 clinical patients. I don’t believe that the Winter and Hinman articles were wrong. It doesn’t take rocket science to understand why a scab might be hindering wound healing progress. But I also don’t believe that the Cochrane findings have to be dismissed because they don’t fit our view on today’s wound care. It does mean we have a gap in our knowledge. The gap can be summarised as we are not able to explain why Cochrane and Winter are both right.

Interesting is that all stakeholders in wound care should be aware of this gap since there have been no repeating studies which proof the moist paradigm in different circumstances.

For some reason we have chosen to ignore this issue for the last 50 years which sadly also lead to 50 years of “less useful” research which papers invariably end with the conclusion “more research is needed”.

To me it appears we have some work to do.

To paraphrase: if not me, who… if not now, when.