MD, PhD, FMedSci, FRSB, FRCP, FRCPEd.

Across the world, researchers are frantically trying to find a treatment for COVID-19. Thus many trials have been initiated – and some are better than others.

The aim of this study was to develop a new Chinese medicine (CM)-based drug and to evaluate its safety and effect for suppressing acute respiratory distress syndrome (ARDS) in COVID-19 patients.

A putative ARDS-suppressing drug Keguan-1 was evaluated in a randomized, controlled two-arm trial. The two groups were:

  1. A control therapy receiving alpha interferon inhalation, 50 µg twice daily; and lopinavir/ritonavir, 400 and 100 mg twice daily, respectively.
  2. The experimental group receiving the control therapy plus Keguan-1 19.4 g twice daily. The new formula was derived from 3 different formulae, Yinqiao Powder (银翘散), Sangju Drink (桑菊饮), and Sanren Decoction (三仁汤), named “Keguan-1” (meaning anti-coronavirus 1 in Chinese) with 7 components: Lonicera japonica Thunb. (Jinyinhua, lot. 19040301) 30 g, Forsythia suspensa (Thunb.) Vahl, (Lianqiao, lot. 19040221) 30 g, Morus alba L. (Sangye, lot. 19045321) 15 g, Chrysanthemum morifolium Ramat. (Juhua, lot. 19040811) 10 g,
    Coix lacryma-jobi L. var. mayuen (Roman.) Stapf, Yiyiren, lot. 19025161) 30 g, Fritillaria thunbergii Miq. (Zhebeimu, lot. 19041161) 15 g, and Prunus armeniaca L. var. ansu Maxim. (Kuxingren, lot. 19045591) 9 g. The powder versions of the drugs for the 7 components of Keguan-1 were obtained from Beijing Tcmages Pharmaceutical Co. Ltd. (Beijing, China) and mixed in the defined ratio.

After 2-week treatment, adverse events, time to fever resolution, ARDS development, and lung injury on newly diagnosed COVID-19 patients were assessed.

An analysis of the data from the first 30 participants showed that the control arm and the testing arm did not exhibit any significant differences in terms of adverse events. Based on this result, the study was expanded to include a total of 48 participants (24 cases each arm).

The results show that, compared with the control arm, the experimental group exhibited a significant improvement in time to fever resolution (P=0.035), and a significant reduction in the development of ARDS (P=0.048).

The authors (who almost out-number the patients in the study) concluded that Keguan-1-based integrative therapy was safe and superior to the standard therapy in suppressing the development of ARDS in COVID-19 patients.

The mixture used in this trial is adventurous, to put it mildly (although the authors state that they report the development of a TCM-medicine “We have reported here the development of Keguan-1, a new CM drug that was specifically designed for suppressing ARDS development in patients of COVID-19 and/or of other RDI” they actually do nothing of the sort). The trial design is flimsy, to put it politely. And the conclusion is unjustified, to put it scientifically.

One might even say that – pandemic or not – it is irresponsible to conclude from a sample size of 2 x 24 on the safety and efficacy of any therapy – TCM or not.

 

14 Responses to The 1st randomised clinical trial of a Chinese herbal medicine for COVID-19

  • Even if we assume perfect execution, no errors with the data, no fraud, etc, this kind of small-scale study is bound to throw up spurious false postive outcomes. We have no idea how many other trials on TCM ‘medicines’ for COVID have been conducted, and had a negative outcome. They won’t get published. So, these studies pollute the medical literature, and sow mistruths and misinformation – all to the detriment of effective healthcare. The only justifiable reason for running a small scale trial for a new putative treatment is where there exist strong reasons to believe that the treatment offers a decent chance of an important therapeutic effect. This means that good scientific grounds must exist to indicate a decent chance of success – these grounds include in vitro studies and animal studies, and -most importantly- that there exists robust theoretical support for the likely biological effect of the treatment, in terms of our understanding of physiology, pharmacology, biochemistry, immunology, etc. In other words, it is crucial that the proposed therapy is PLAUSIBLE. Otherwise, one could run individual trials of completely preposterous ‘therapies’ -e.g. the effect of wearing tartan underpants on COVID severity- and a few of these trials would inevitably, entirely by chance, throw up a false postive result – with ‘statistical significance’ attained – namely that wearing tartan underpants reduces the risk of COVID patients developing severe disease. This means that it is unethical to conduct a clinical trial on therapies that lack plausibility. To conduct such studies can only create confusion and error in the field – and this is important, because it can translate into treating patients receiving ineffective medicine, and not receiving the effective treatment. (Another reason that such trials are unethical is that by recruiting patients to a useless/potentially misleading study wastes their time and risks their health.) Looking at the TCM paper that Edzard reviews here, what we have is a bizarre concoction of botantic type ingredients that, somehow, the researchers hope might have a bbeneficial effect in COVID patients. Where’s the plausibility in that? I fail to see it. If it’s not there, then this research is UNETHICAL.

    • hear, hear!

    • Oh yes, “plausability” is certainly much more important than evidence. Really? Do you really believe that stupidity?

      As for your other complaints about this study, doesn’t having it randomized obliterate your concerns? Wouldn’t there be as many false positive results on BOTH sides of this trial?

      And Ernie…you refer to this study as “sloppy,” but it is so unlike you to avoid giving specifics. Doesn’t this suggest that your review of this study is sloppy?

      And although this trial was small, wouldn’t THIS mean that it can only be possible for the results to be statistically significant IF the differences between the treatment groups be LARGE?

      It is so interesting watching how you and your fellow cultists analyze trials by what you say and by what you don’t say.

      • you’ve been smoking something interesting, Dullman?
        I never said it was sloppy!

      • in case you mean ‘flimsy’ – yes, is a sample size of 2×24 not concrete enough for you?

      • Dear Dana,

        there is a German cabaret artist and comedian who said a very clever sentence years ago: “When you have no idea: Just shut up.”

        You should follow this advice, it works wonders.

        • Oh…what a sophisticated response!

          Telling someone to shut up.

          Oh…and that works so wonderfully well.

          You are SO evidence-based. NOT.

          My personal condolences…

          • Dana. You said:
            “ And although this trial was small, wouldn’t THIS mean that it can only be possible for the results to be statistically significant IF the differences between the treatment groups be LARGE?”
            This reveals that you have a fundamental misunderstanding of clinical trials and statistics. If you were correct, then P-hacking and the reproducibility crisis in medical (and other) research would not exist. But these problems certainly DO exist, and are responsible for the spread of much misinformation about ‘what works’ in medicine – particularly in CAM.
            Respectfully, you need to read a lot more about research and statistics, before making such comments. Otherwise, you can hardly complain if you get comments along the lines of ‘shut up’ from those who DO know what they’re talking about.

          • Dear Dana, have I hit a sore spot? I am so sorry (not).

            You should team up with Dr. Hümmer. He is also always in a bad bood, if somone points out his ignorance.

      • Oh look. It’s Dana. Back to display his stupidity on a public platform. Again.

        Yes, Dana. Plausibility is everything. Extraordinary claims require extraordinary proofs. A unicorn walked past my house the other day. You want proof? I’ll send you a picture of the unicorn shit in the road. You say it’s just some horse shit that I found there? Well it isn’t. It’s unicorn shit. Look. It’s got rainbow sparkles on it. That proves it.

        Run along, Dana.

  • Apart from basic mistakes like making multiple comparisons and adding patients until you get p < 0.05, this paper makes the usual mistake of confusing p values with the false positive risk. A p value of 0.035 would correspond to a false positive risk of around 20% even if you were willing to give the concoction a 50:50 prior chance of working. And it would be a great deal bigger if you took into account the implausibility.
    See https://www.tandfonline.com/doi/full/10.1080/00031305.2018.1529622

  • It is hard to conclude very much with these numbers other than that it might be worth conducting a larger trial.

    Although 48 patients were recruited (and 47 were analysed), the primary end-point was the development of adult respiratory distress syndrome (ARDS) and only seven subjects were affected.

    I was also rather concerned that the terms “mean” and “median” seem to be used interchangeably by the authors, which really doesn’t give me any confidence in their grasp of statistics.

    The control group were also older, heavier and had had symptoms for longer before they were enrolled in the trial (at which point everybody, controls and interventional subjects, started standard treatment.

    I can’t work out from the paper what happened to the person who dropped out. Assuming that the outcome was known, they should have been included in the analysis on an “intent to treat” basis, i.e. according to which treatment they were randomised to, not which treatment they received. This may seem odd to non-statisticians, but although it dilutes any effect it also reduces bias, which is more important.

    The authors report their findings in terms of statistical significance in several places, although this has largely fallen out of favour for studies of this type as it can be quite misleading (confidence intervals are better, although often misunderstood).

    This is definitely not a study on which to base clinical practice, but it could form the basis for a better-organised trial with larger numbers and the input of a statistician. If the findings are confirmed then it would be interesting to try to isolate the pharmaceutical substances responsible for an effect.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Recent Comments

Note that comments can be edited for up to five minutes after they are first submitted but you must tick the box: “Save my name, email, and website in this browser for the next time I comment.”

The most recent comments from all posts can be seen here.

Archives
Categories