Lyumjev part 4. Persistence is king…

As some readers may have seen, I was somewhat disappointed when I first put Lyumjev into a pump and started using it with a DIYAPS system. To such an extent, that I decided to mix it with Fiasp to see if that would help… Especially given the numbers of people that were successfully using Lyumjev without any issues.

After nearly a month of the 1:1 Lyumjev:Fiasp mix, I found myself with unequal amounts of Fiasp and Lyumjev at a point in time where I needed to reload the reservoir, so decided that I’d use what I had and moved to 3:1 Lyumjev:Fiasp. I was expecting this to cause me similar issues to the previous occasion with full Lyumjev and was quite surprised when this didn’t happen.

As a result, I decided to change over fully to 100% Lyumjev and see how we go on with that in the pump and OpenAPS together. I also elected to use a stainless steel infusion set on the basis it may cause few reaction issues. And I have to admit that I’ve been very pleasantly surprised.

Not only am I getting the fast action of the Lyumjev, the site issues that I discussed after using it for a week seem to have just disappeared. In fact, the performance is really quite impressive.

As we can see from the above 12 hour image, with 100% Lyumjev, there appears to be a similar result as with Fiasp in relation to announced meals. It works pretty quickly and brings you back in line without a huge rise.

What I found to be far more interesting was the eating without announcing, and not bolusing. Tested on a few occasions with 50g or so of carbs, there seems to be a roughly 4 mmol/l rise, which the system keeps in check, followed by a slow decline. Admittedly, the case shown here I then applied an extra carb load and we continued higher, but with a peak of 11.1mmol/l, I was very impressed with the outcome.

But what about the site?

Some of the issues that I had first time around and that I know others have experienced include pain on bolusing and soreness at the site.

The first of these happens occasionally, but not very often, and tends to be felt early-ish after a set change. The second seems to have stopped. It no longer feels like a massive bruise forming at the site. I also switched back to Teflon infusion sets to see if that made any difference, and I’m pleased to report that there were no adverse side effects.

Again, demonstrating that this seems to work well, after a high temp target from a cycle ride followed by a thai meal with underestimated carbs, the following happened:

In this case, we can see the glucose level climbing after a small amount of carbs then dropping as a result of the cycle ride, before eating with no bolus with the meal. The Lyumjev has acted as fast as it is expected to, with the additional benefit of exercise to add to that.

The other effect that seems to appear here is the stability in glucose levels after the dosing is complete. It’s not 100% clear whether the tail of the insulin is really five hours, indeed, I am using it at seven hours and that seems to work quite nicely, but the way the reduction in the tail works seems to be noticeably less violent than Fiasp can sometimes be.

But why has Lyumjev suddenly started to work, with no site issues to speak of? 

So what’s going on?

After my initial use and the decision to revert to a form of mix, I had conversations with a pharmacist and an endocrinologist.

Both provided some insight into Treprostinil that I was unaware of.

  • Those who use it for pulmonary hypertension often wear sets for 30 days, which may look bruised but don’t have large amounts of pain;
  • These are the same sets we use for insulin;
  • It often requires acclimatization to the use of Treprostinil, with new users taking a number of sets or a period of time to have it work well.

That’s some food for thought.

I’ve now had this new insulin for around 2.5 months and have started a fourth vial. I’ve also changed the way that I use the insulin, manually bolusing only small amounts and relying on SMB for everything else. That’s a little different to what I was doing in my first week of use where I was taking larger doses.

Additionally, this is a second batch from the pharmacist. Taking all of these factors into account raises some interesting questions.

The first and most obvious of these is whether the issue was related to the batch of Lyumjev I had received from the pharmacy. I noted that it didn’t seem to work at all well in the pump, however, it did work with injections, so I suspect that this isn’t the case. It’s always a possibility though.

Secondly, based on the previous conversation points, has it taken me this long to acclimatise to the additives in this new insulin? As per the commentary that Treprostinil can take a while to be used effectively in its primary setting, maybe for some of us, this also applies when taken as part of an insulin concoction. If this is the case, using it in a mix with any other insulin potentially might reduce the exposure level to the additive and allow the body to get used to it. It’s worth noting that in the studies that I can find, pain on first use was shown to be common in 80% of users and the acclimatization when used for pulmonary hypertension can take months, and does not happen for all.

Finally, could it be the dosing technique causing the issues? By reducing the size of a single bolus delivery, do I limit the amount of additives per bolus to a level that works with my body and stops the unexpected lack of effects from taking place?

There we have three possible theories as to what has happened with Lyumjev. One thing that is intrinsically clear is that it now works, and that’s something I’m very happy about. 

It also suggests that there may be hope for others that have had issues with Lyumjev. Maybe, with a bit of persistence, and a change to how you deliver insulin, it’s possible to use it effectively. The only way to truly know is to give it a shot!

As the saying goes, if at first you don’t succeed, try, try again….


  1. Great development, congrats! I think you are on the right path …
    I am still doing 100% UAM with LJ und AAPS 2.7 with the great results (97% TIR, 6.0 A1c estimate, GVI 1.2) … now doing it for five weeks without a break!

  2. Thank you for your persistence in working with this insulin and automatic system. What concerns me it that this insulin and automatic system may encourage the eat what you like crowd of TIDs to just let the system cover their largesse. We know from US TID Exchange that in most age groups of TIDs obesity and overweight rates are increasing. Dr Hamby of Joslin Diabetes Centre opines that this trend is related to to the eat what you like and dose for it education programs.

    • Tony, in both of your responses, you’ve highlighted that you are concerned about growing levels of obesity amongst those with T1D that have access to automated systems and faster insulins.

      I don’t think this is any different from the population in general, and those who opt to eat what they like will do it anyway, so having better outcomes and lower complication rates by providing better treatment (where it can be accessed) must be the right way to go. Those who want to eat what they like tend to do so anyway.

  3. Hi, I’ve been using 75% bolus since the beginning and I let SMB do the rest and for snacks I let UAM handle it. I can’t do without bolus, probably because of the resistance. Since Lyumjev I have also had severe burning and hardening, as I use the DanaRS I set the bolus delivery to 30 s/rev and the side effects disappeared. Over time, the excess insulin consumption also decreased. Now I do not want to change

  4. Hi Tim, I found this extremely helpful. Are you still using Lyumjev with success? I am on day 2 but with good results so far with very, very mild pain at the infusion site if at all.

    On the treprostinil point, it appears that the amount used in Lyumjev is dramatically smaller than the amount used by those with pulmonary hypertension (like, it looks like they use more in an hour than most of us would use in a week, though there is a wide range of doses and I could easily have been an order of magnitude off). That might explain the shorter acclimitization time.

  5. Always interesting to read about your n=1 research!

    Perhaps you could also look into the effect of larger boluses, now that you have found Lyumjev to work for you. Perhaps not so tempting to change your current strategy, but it should be fairly quick to find out if it is the bolus size that matters in your case.

    We don’t have access to Lyumjev just yet, but it is very nice to be able to learn from your experiences while we wait!

    “Finally, could it be the dosing technique causing the issues? By reducing the size of a single bolus delivery, do I limit the amount of additives per bolus to a level that works with my body and stops the unexpected lack of effects from taking place?”

Leave a Reply

Your email address will not be published.