The patients were diagnosed as having POTS so at some point in the past they must have done a tilt table test with the HR increase and have the corresponding OI symptoms. I agree the 30bpm is somewhat arbitrary, all sorts of factors depending on the type of test, length of test, hydration levels... would all impact the results. But that is a separate discussion about how to best group the people with the symptoms of POTS and what kind of role the Tachycardia plays. I don't think that really makes any difference in regards to the results of this trial.
Randomized Trial of Ivabradine in Patients With Hyperadrenergic [POTS], 2021, Taub et al
ChronicallyOverIt
Senior Member (Voting Rights)
I’d say a huge reason for this is it’s damn hard to feel good or even concentrate when your heart is at 130-140bpm just from standing. Makes it near impossible to focus on anything. Get up to make a sandwich for lunch feels like you’re in a marathon. Going for a walk with a friend and want to socially chit chat, hard to engage when your chest is pounding and your body is screaming.
Subjectively you can be more present if you don’t feel like there’s a bomb going off in your chest. These results make sense from my lived experience of unmedicated POTS.
As far as I can see, the criteria for the people in this study did not include a requirement about a heart rate increase upon standing. I also quoted the table (Table 2) that gave the baseline and placebo changes in heart rate recorded in the study. The means were less than 30 bpm. The participants did not have POTS as I would define it at the time of the study.
There's no indication that these people had heart rates as high as that. The baseline heart rate following standing was only 95 bpm +-16.8. Have a look at the Table 2 that I quoted upthread. I think many healthy people would have a similar maximum heart rate during a 10 minute stand.
The people that this study investigated did not meet the usual definition of POTS.
Last edited:
POTS is a clinical disorder classified by: 1) symptoms upon standing such as lightheadedness, palpitations, tremors, weakness, blurry vision, and fatigue; 2) increase in heart rate ≥30 beats/min upon postural change from recumbent to upright position within 10 min of standing; and 3) absence of orthostatic hypotension (3,5,8,30). Hyperadrenergic POTS, a subtype of POTS, is defined as an elevation in NE >600 pg/ml upon standing and a systolic BP increase of >10 mm Hg when standing upright for 10 min (5). We did not use the systolic BP criteria, as we recognized clinically that many patients had overlapping subtypes (e.g., hyperadrenergic and hypovolemic that resulted in lack of systolic BP increase). A positive head-up tilt table test (HUTT) (heart rate ≥30 beats/min) and NE (≥600 pg/ml) were required for study enrollment.
These issues were also addressed in the following editorial: https://www.jacc.org/doi/10.1016/j.jacc.2020.12.028.
I would also point towards the following press release: https://ma1.mdedge.com/content/ivabradine-knocks-down-heart-rate-symptoms-pots and the ACC summary: https://www.acc.org/Latest-in-Cardiology/Clinical-Trials/2021/02/16/19/13/Ivabradine-in-POTS.
I would also point towards the following press release: https://ma1.mdedge.com/content/ivabradine-knocks-down-heart-rate-symptoms-pots and the ACC summary: https://www.acc.org/Latest-in-Cardiology/Clinical-Trials/2021/02/16/19/13/Ivabradine-in-POTS.
The editorial says
While I disagree with some of what the editorial says, I basically agree with them on that sentence.
I did see the head up tilt table requirement. But, there is a big difference between a standing test for 10 minutes and a head-up tilt table test. A requirement of >30 bpm in the standing test is quite different to the same requirement in a HUTT. I have seen discussion querying how diagnostic the HUTT is - people with no orthostatic symptoms can qualify for a POTS diagnosis in the HUTT.
I think some of the reasons for concern have not been well founded as already discussed by others (both the blinding argument and the seemingly circular argument on objective measures). However, I think the following looks like a rather genuine reason for concern: Apparently there was a different study by the same authors on POTS (rather then hyperadrenegic POTS) that was completed which however never reported results: https://clinicaltrials.gov/study/NCT03182725?term=Pam Taub&rank=6.
In my view the evidence for there being different subtypes of POTS is essentially non-existent and from what I've seen rather boils down to post-hoc story telling. If the trial for POTS is negative then its really hard to understand what to make of the above data. There are currently larger trials for POTS with Ivabradine as part of RECOVER for Long-Covid of which the above authors are part of. I don't expect positive results but I also don't expect well run or even sensible studies.
In my view the evidence for there being different subtypes of POTS is essentially non-existent and from what I've seen rather boils down to post-hoc story telling. If the trial for POTS is negative then its really hard to understand what to make of the above data. There are currently larger trials for POTS with Ivabradine as part of RECOVER for Long-Covid of which the above authors are part of. I don't expect positive results but I also don't expect well run or even sensible studies.
Last edited:
The most obvious objective outcome measure would surely be time spent upright throughout the trial, since the main disability of POTS is not being able to be upright.
I see the participants all had to have HR increase over 30 during TTT, so that could be tested again while on the drug or placebo.
I see the participants all had to have HR increase over 30 during TTT, so that could be tested again while on the drug or placebo.
I think that is this study. If you go to the study protocol you find:
"The clinical trial will be registered under www.clinicaltrials.gov. Subjects between the ages of 18-65 who have a POTS diagnosis will be enrolled. Specifically, we will enroll patients who have hyperadrenergic POTS as confirmed by norepinephrine (NE) levels > 600 pg/ml. This is a 3-month study with a randomized double-blind placebo controlled cross over design."
The study also concludes about the exact same time as that paper was published.
I agree with you on the subtypes of POTS. I think anything other than defining POTS as a syndrome with certain OI symptoms associated with a higher than normal increase in HR on average is asking for trouble.
I think you're right. Looks like they listed the study more than once and I didn't do enough reading. I edit my comment above.
Who says this applies to patients in the trial? I'm not too sure. If you simply have patients that have POTS in your trial it can be quite hard to make out what their problem is. Some might have brainfog combined with an increase on a HUTT for others it might dyspnoea during activity (+HUTT positive) for others its headaches (+HUTT positive) and for others its insomnia (+HUTT positive) which by definition never occurs when upright. I don't think we really know what the problems in this trial was for patients. I think its entirely possibly for someone to have insomnia, migraine and exercise intolerance or alternatively tremors, nausea and coat hanger pain and be given a POTS diagnosis as a result of this.
Repetition of the TTT seems wise but the results already indicate that these people don't have abnormal increases of heart rate upon standing, so it's a bit hard to make out what it would mean. As @Hutan commented a 16 BPM increase as seen in this study upon standing is very much in the normal range. What do the TTT results tell you then even if there are group differences?
Last edited:
Utsikt
Senior Member (Voting Rights)
Yes, and that’s my point!
The only purpose of using blinding of participants is to eliminate the risk of bias caused by participants knowing which intervention they have been assigned to.
If the blinding doesn’t work because the participant for whatever reason can notice when they are on the active drug, then subjective outcomes become unreliable in the same way that they would when you combine open-label and subjective outcomes: due to the substantial risk of bias.
I have not hand waved them away, I have said that broken blinding makes the subjective outcomes unreliable. I’ve also explained why the objective outcomes do not really support that the treatment is effective, for different reasons.
Jonathan Edwards
Senior Member (Voting Rights)
I do think this is problematic. It seems likely that ivabradine reduces tachycardia. Beyond that we don't know whether any subjective reports of other aspects of QOL are attributable to that reduction in tachycardia (which is quite plausible) or to unblinded role-playing on the part of patients who know they are supposed to be appreciative.
The interpretation might depend on whether they knew that the drug was supposed to reduce tachycardia directly or make them well for unspecified reasons.
I think a significant problem here is knowing what this population really is. Why were 20 out of 21 women? Are these people who present to cardiologists because they are bothered by palpitations rather than OI per se and therefore a very different group from people with OI as part of a wider spectrum of problems who happen to test positive on the 30bpm criterion?
I think the easy and straightforward thing to have done first would have been to assess blinding directly as @Hutan already pointed out by asking participants whether they could guess their allocation arm. Then subsequent discussions could have been useful. Otherwise one is left discussing things one cannot know about a population one has no idea about. In the situation where people improving are more likely to think they got the drug even this data wouldn't tell us too much.
The graph of the study does suggest the possibility of people guessing whether they received treatment or not sufficiently often to drive results: Everyone in the placebo-group opted for the drug at crossover but only 60% of people in the drug group wanted the placebo. Supposedly this shouldn't matter as the majority of withdrawals are due to treatment side-effects, however all of these seem to only effect one group which is maybe a bit more than one would expect purely by chance. Of course it's possible that this is just a random phenomenon. We would know if the authors had provided useful data.
The graph of the study does suggest the possibility of people guessing whether they received treatment or not sufficiently often to drive results: Everyone in the placebo-group opted for the drug at crossover but only 60% of people in the drug group wanted the placebo. Supposedly this shouldn't matter as the majority of withdrawals are due to treatment side-effects, however all of these seem to only effect one group which is maybe a bit more than one would expect purely by chance. Of course it's possible that this is just a random phenomenon. We would know if the authors had provided useful data.
Last edited:
Utsikt
Senior Member (Voting Rights)
I’m leaning towards not giving the authors the benefit of the doubt, because that would mean that it’s ‘profitable’ for them to withhold information. I usually assume it went wrong if they have not proven it didn’t. After all, they have the burden of proof.
As far as I'm aware assessing blinding via asking participants about guessing their allocation arms is not standard. Possibly because it doesn't tell you anything in situations where people guess they are on the treatment because they are getting better, rather than the blinding not working, or because one is worried that such questions alter behaviour, I don't know. Thinking about it more, in the above situation of the trial I probably overestimated its use in my previous post. I find it reasonable that this data wouldn't have told us anything about the blinding rather than other effects. I guess you're gonna have to do things differently from the beginning rather than just adding one set of questions.
(I think the situation for Rituximab or Daratumumab could be rather different because there you can ask people before they can perceive any benefit which doesn't seem like a possibility for the above trial).
Last edited:
Utsikt
Senior Member (Voting Rights)
To be blunt, I don’t care about what’s standard. Lots of standards are insufficient.
I don’t follow this. If they can guess if they got the active treatment, the blinding isn’t working. For this specific matter it doesn’t matter why they could guess it.
I know that it has been proposed that asking such questions might alter behaviour, but I have a very hard time imagining that participants would not already be trying to figure out which group they were in.
An easy solution would be to not mention anything about it until the very end after all of the other data has been collected.
No. Your blinding could be working perfectly if people are more likely to guess they are receiving active treatment whenever it works. That doesn't necessarily mean the blinding didn't work. I think what then matters is what the reasons for your guessing are, at what point those questions came, whether the investigators are also guessing correctly ect. If correct guessing is due to perceived efficacy then that doesn't necessarily mean blinding was destroyed. It could be fully in tact. Correcting guessing and unblinding are not the same.
The following is an example: Think of the situation of a 100% effective treatment for bumps on the head that works after exactly 80 days. If the correct guesses are roughly equally distributed at day 79 then blinding works even if people then change their guesses once they see the effects on day 80 and in this setting you managed to control for all sorts of variables that you want to control.
It becomes a bit more complicated if all results are subjective as seen in the above trial. I agree that in the above situation results can be driven by anything without knowing what that could be.
Last edited:
rvallee
Senior Member (Voting Rights)
I really can't follow the arguments about blinding having been broken because it makes a noticeable difference, they make no sense to me.
That can't be how it works, because any effective treatment achieves this. In fact, the best case scenario for a clinical trial is when the improvements are so noticeable that it can't be considered ethical not to give the active treatment to the control group, ending it early. Clinical trials are very messy things, about as far as lab experiments as it gets.
The differences in heart rate are definitely significant, impressive, even. Whether they represent a meaningful benefit is another question, although from experience, that reduction alone is a clear benefit in itself, but there is definitely an objective improvement, and for sure the drops in heart rate are the kind that is noticeable when you are used to the higher, unstable, heart rates.
There remain some problems, but that Ivabradine lowers excessive heart rates in this patient population is clear, and the fact that they don't meet POTS requirement is a problem for "this applies to POTS", rather an odd choice, but it still shows it can produce a significant reduction, even to people who don't meet the arbitrary threshold, which is useful information in itself. Having lived with meeting POTS criteria for months, I can definitely understand why it's hard to find such participants, for similar reasons why trials with severe ME/CFS patients require a lot of extra work.
Last edited: