I’m a diabetic. Unlike some people who seem to take great pleasure in identifying themselves and building personal brands as diabetics, I don’t like talking about my diabetes. For more than 30 years I’ve worked hard to minimize others’ association between myself and the disease, hopefully with a modicum of success.
I’m going to step outside myself this once to indulge in a rant about the state of diabetic care when it comes to pump therapy. This may come off as too long, too angry, and too frustrated to many people, so let me get the too-long/didn’t-read out of the way first (allow me to adjust my tinfoil hat for the moment):
TL;DR: Companies who create diabetic therapy devices and treatments do not want diabetics to get healthy or, especially, find a cure. They want to create long-term dependencies and actively make treatment more difficult over time to ensure that outcome benefits them, not the patients.
I have been an insulin pump user for 15 years. For a while, the pump helped with the blood sugars, but there were distinct trade-offs. For example, if you forgot whether you took your insulin you could check the record and see when the last time you gave a dosage. However, you had to wear the pump 24/7/365 with no break, which meant that you can kiss a good night’s sleep goodbye. You didn’t have to deal with needles, but you would build up scar tissue at the setting sites.
Things of that nature.
Nevertheless, on the whole the simplicity of the dosage routine, and the reliability of the pumps to provide that dosage, made all of the grief “worth it.”
In the late 2000s, Medtronic (the only pump manufacturer in the US at the time) came out with a Continuous Glucose Monitoring System (CGMS). It was the first real attempt to create a feedback system from your own body back to the pump to modulate insulin flow. In other words, it was the first attempt to approximate the functionality of the human pancreas.
It was an unmitigated disaster.
The sensor needed to be inserted at a sharp angle into the skin, and locked in place. Any movement of the sensor catheter sent the insulin pump into a screaming alarm and forced a reset.
Oh, and if it happened more than a handful of times, that reset had to be done at the factory. Yup. Fan-freakin’-tastic.
For me, this could happen just from natural breathing movements. I tried it for a week, never slept for one single minute, and spent the entire week focusing on the pump and diabetes rather than anything else (including work, sleep, anything).
The experience was so traumatizing that I vowed never to use a CGMS that required such a strong tether to the pump ever again.
The 5 year plan
Insulin pumps are very expensive. The starting price for these little suckers is $6000. When you get a pump, the insurance (that I have) will cover the pump for 5 years, even though the warranty for the pump is only 4 years. This leaves a wee little gap, doesn’t it?
Sure enough, cast your mind back to October, 2015. My Medtronic Paradigm pump, EOL’d years ago, was out of warranty and I am due for a replacement in January, 2016. Medtronic was kind enough to offer me a 90-day loaner of the same model while I did my homework and found a suitable replacement.
Doing My Homework
Since I’m going to be stuck with any pump that I get for about 5 years, I wanted to make sure that I didn’t miss up any opportunities. A lot can happen in 5 years and I wanted to see what kind of advances in technology were available.
At the same time, I had some very specific criteria. Given my nightmare of an experience with the CGMS earlier, I have been wary of the technology and refuse to sign on to a pump system that requires them.
Equally as importantly, the pump has to support me, not the other way around. I do not simply want to be a walking diabetes medical display unit.
One of the key criteria that I had was a simple one. I travel a lot: at times I can be on the road for work up to 80% of the time. This means I get hassled by the TSA nearly every time I fly, so giving them something else to screw with doesn’t work for me. I’ve had TSA agents rip out insulin settings before (with alarming frequency, in fact), and I’ve lost count of the number of times something has gone wrong on an airplane and I needed to resort to an emergency backup setting.
So, at the top of my “lifestyle” criteria was one very simple item:
Any pump needs to be able to have the site change in under 60 seconds in an airplane bathroom.
Note: I can do this now with the pump that is now end-of-life’d. You’ll see below just how laughable a requirement this actually is.
Other criteria include:
- Secure. Pump could not be remotely accessed or manipulated by unauthorized/malicious individuals
- Accurate/Reliable. Therapy could not inject wide variance of reliability into insulin delivery
- Rapid Setup/Reload. Due to demanding travel schedules, pump setting exchanges must be accomplished in 60 seconds or less
- Non-Dependency of Auxiliary Equipment. Primary functionality of insulin pump should not be compromised by interconnectivity requirements to ancillary technologies, gadgets, software; or dependent upon the efficacy of other technologies, gadgets, or software (e.g., CGMS)
- Non-interference. Pump therapy could not be more intrusive than existing Paradigm requirements; higher frequency of setting changes, slower priming, extra materiel and clutter were all to be avoided. The pump (and related system) should be discrete, silent, and not draw attention
- Insurance Approval. Solutions must be covered by my insurance policy
Let’s take a quick look at that first item, shall we?
In 2011, controversy erupted when it emerged that the Medtronic Minimed pumps (which is the one that I have) can be remotely hacked. This was revealed at the Black Hat security conference in 2011, where it was shown that hackers could remotely access pumps from up to 300 feet away.
See, Medtronic’s remote communications occurs unencrypted (even today, according to conversations I’ve had with Medtronic in the last month). The researcher was able to break the system with a $10 radio frequency circuit board.
Medtronic’s reaction was appalling. After initially agreeing to work with Jay Radcliffe, the security researcher, claiming that they take this “very seriously,” it appeared they waited for the fervor to die down and weather out the storm. They did this, despite knowing about these types of flaws in their systems (they also make pacemakers) since 2008.
Medtronic argues that you need to have the serial number of the pump in order to make the attack work. There argument is that it is virtually impossible to have physical access to the pump without the users’ knowledge to obtain that information. However, as Radcliffe pointed out, this is “security by obfuscation,” which of course means that it’s not security at all.
Even so, Medtronic has a point. It’s not like someone is going to go and be able to lift the serial number off the pump still attached to someone’s body, right? So what could someone do to get those serial numbers? Let’s see… what could someone do… what could possibly put people at risk…
How about a Medtronic security breach where patient records are compromised? Yeah, come to think of it, that might just do it.
I thought I was being reasonable. All I wanted to know was how Medtronic had addressed the problem, and was just looking for reassurance that there was nothing to worry about. However, when I did so I found no such assurance. In fact, what happened made me more suspicious and concerned than ever.
When I spoke to Medtronic about this, I was told that “no one” had ever asked about this before. I find this slightly plausible, in the most technical interpretation of the term “plausible.” I also found the response considerably patronizing. I was asking about it. The news reports are four years old, and I would have thought that they would have some sort of solution by now – or at least some type explanation to show that their communication systems are secure. With the interconnectivity between insulin pumps and CGM systems, it’s not simply a matter of “knowing the serial numbers” – now that data can be sniffed over the air, and I wanted to know about the software hardening.
The point is that it doesn’t matter if no one else has asked about it. A medical equipment manufacturer should have the answer to this question “ready to go.”
But, who am I, right? I mean, if I don’t buy a Medtronic pump, I’m just one small pebble in the ocean.
Or maybe I’m being too harsh. After all, in 15 years, I’ve never been hacked (even though it takes only once, but hey, trying to be fair here). I should take a look at the probable risk versus the actual risk, right?
Not Much Competition
Medtronic Revel; 530G with Enlite
As I mentioned, the version of the pump that I have now is a Medtronic Minimed Paradigm pump (I often use the terms “Medtronic” and “Minimed” interchangeably). As that pump is no longer available, I started looking at the two options that they have available.
The Revel, an earlier model evolved from the Paradigm system, has the option to have CGMS integration but is not a requirement, fortunately. Various UI improvements over the Paradigm model gave the Revel positive reviews upon its release.
Medtronic’s latest system however, 530G with Enlite, is a combination Pump/CGMS) that cannot be disaggregated. Reviews of the Enlite system, a successor to the horrible CGMS system that I tried before, have been rated “mildly negative” to “poor,” and as it is a permanent requirement of the 530G, made the option inappropriate.
Moreover, the 530G system requires more blood sugar measurements (5+ times/day) with traditional methods in order to maintain calibration with the Enlite CGMS, negating the usefulness of the CGMS altogether. So, why bother with a CGMS if you have to test your blood sugar as much or more than without?
“For your convenience,” of course.
Now, perhaps I’m being a bit unfair here. One of the key selling features of the CGMS/pump combo is that it has trending capabilities: theoretical predictive capabilities that can help avoid highs or lows, and help warn users about “double dosing” to correct those highs or lows. This is an advantage, and should be identified as such.
The system, therefore, requires three bulky devices to be carried at all times (CGMS, pump, blood sugar monitor), as well as all ancillary materiel for all three devices. As the devices must be used in conjunction with one another, the configuration/setup time can be prohibitively long, which means an impossibility of making changes during travel scenarios (e.g., on an airplane).
Medtronic calls these devices “small and unobtrusive,” but the reality is anything but.
First, take a look at where the sensor is supposed to be inserted:
This doesn’t leave you with an awful lot of options, and some of them are somewhat awkward to reach. Then there’s the amount of real-estate on the body it takes up once it’s inserted:
This is not to mention all the equipment you have to carry with you at all times:
And this is just the stuff to carry around for the glucose monitor.
Even after all this, Medtronic’s biggest issue is that the pumps are still wildly insecure. Now, my conversations with the company were limited to ordering personnel who were not technical experts, but they stated they could not forward the inquiry to sales or engineering personnel, and recommended that the sales representative be contacted. When I did reach the sales contact, she told me that no one had ever asked the question about security before. Considering she covers the highly technical area of SillyCon Valley, I can’t help but find this somewhat difficult to believe.
Even so, of the insulin pumps that match most of the criteria, the Medtronic Revel came closest to matching, but the insecure nature eliminated all Medtronic devices from consideration at this time.
Tandem: t:slim; t:flex
Despite apparent reports in popularity for the Tandem products, the two devices (distinguished primarily by reservoir capacity) suffer greatly from convenience and design flaws.
Notably, reservoir cartridges can take up to 15 minutes to prime, require near-perfect “gravitational alignment” to avoid air-bubbles, and tethered charging up to two times a week.
Let me reiterate that key point: Up to 15 minutes to change a setting.
One youtube diabetic did an excellent breakdown of the changing of the Tandem pump and, even allowing for the necessary time for instruction, her demonstration took over 12 minutes.
However, she didn’t go through the process of trying to ensure that bubbles don’t form in the infusion lines, which another youtube diabetic demonstrated: even fast-forwarding through several of the steps it still took 12 minutes of video time.
You have got to be kidding me.
Any one of these drawbacks eliminates its usefulness for emergency setting changes on an airplane, for example, and thus were immediately removed from consideration.
As one review put it: “My summary, in a word: ‘It’s pretty, but annoying. And I really don’t trust it.”
Much of the issues surrounding the Tandem pumps strike deep into the main criteria listed above: the air-bubble “clog”, cartridge replacement time (10+ minutes), and “automatic” bolus recalculations, and loud feedback noises. Add to the mix a seemingly endless requirement of button pushes for even the simplest of tasks, the waste of insulin during the priming process, and you get an overall rough experience. All of these disqualify the Tandem from being an effective solution. On top of it all, questionable customer support reports seem to indicate this is a solution to avoid.
Animas: Vibe; Ping
Medtronic is not the only one with a CGMS system: A company called Dexcom has a couple of systems as well, and it appears (at least, anecdotally) that people appear to be liking it. At the moment, it appears that there are two models: G4, and G5, the key difference being (between the two) that the G4 requires another device, to be carried around. It kind of looks like an old iPod model.
Animas insulin pumps tout their integration with Dexcom CGMS so you don’t have to carry around the iPod-looking device; however, the technology is not forward-compatible, and limitations with one create limitations in the other.
For example, new editions of Dexcom software/firmware are incompatible with existing Animas products (which only work with the G4). This means that, should I choose the Animas products I will be forced to use a CGMS that has already been superceded by a later edition. Otherwise, you can use a separate CGMS that’s not integrated with the Animas, but then what’s the point?
The other problem is that the limitations carry over. That is, there is a major limitation in that when you suspend your insulin delivery, the CGM data stream is also put on pause.
Let’s put this into perspective, so that you can get an idea of what this means from a day-to-day perspective.
Most (that is, normal, non-diabetic) people have blood-sugar ranges between 80-120. Diabetics suffer on occasion from low blood sugar, or hypoglycemia. When that happens, it’s not uncommon to suspend the delivery of insulin from the pump so that it doesn’t continue to add to the problem.
However, a common consequence of this is that your blood sugar can skyrocket, and the rate at which it does this can be unpredictable. Therefore, it would help that a “Continuous” monitor would, you know, continuously monitor. I know, I know, call me crazy. What a wild-ass thought.
The Animas, however, shuts off the monitoring along with the pump, completely negating (in my opinion) the purpose for having a CGMS in the first place – being able to monitor blood sugar fluctuations in extreme situations.
The user interface for the Animas is particularly problematic, with rapid screen timeouts, and a <48 hour reservoir capacity, this means that more materiel must be carried at all times (relevant when traveling on a regular basis).
In addition, anecdotal evidence indicates numerous issues in the pump/human interaction. No ability to silence, poor interface feedback and prolonged delays in functionality (see review in Bibliography), and the high likelihood of accidental bolus cancels due to stray button presses indicate poor design and potentially dangerous outcomes.
Let me focus on that last bit just a bit more. When you are giving yourself insulin, any accidental button press cancels the dosage.
Given that the Animas requires a large amount of math to do in your head, or go through the actual calculations on the screen, depending upon when the bolus is cancelled you effectively will have to go through the whole process again from scratch, just to avoid giving yourself more insulin than you intended to give in the first place!
To get an idea of what I mean, here’s a short video of how the Vibe handles correcting a high blood sugar. Keep in mind that this is not fast-forwarded, there are no unnecessary actions in the video, and it still takes over a minute-and-a-half to do a simple corrective bolus:
In addition (yes, there is more), the Animas does not have an “escape” button, which means you better get used to menu after menu. What’s more, if you don’t make a selection in less than 10 seconds, the screen time-outs and you have to start all over from scratch. When you are changing your insulin settings, btw, there are plenty of moments when you can go longer than 10 seconds between button presses.
Ironically, that doesn’t stop the pump from taking it’s own sweet-ass time. From one of the reviews:
Seconds upon seconds are wasted with this pump. You suspend insulin delivery or CGM, or begin an insulin dose, and the Vibe immediately gives you a quick beep or vibration letting you know it’s on Pause. But when you click on Resume, there’s nothing… for a whole 18 seconds, the pump just sits there without changing, locking you onto that highlighted Resume line while you wonder if you’ve actually restarted your insulin dosing and CGM or not. This drove me absolutely nuts.
It would drive me nuts as well. In fact, it would likely send me into fits of rage.
There are other annoyances as well. According to the Animas User Guide, you need the “just right” battery. Use one that’s too weak, you void the warranty. Use one that’s too strong, you void the warranty. Use a rechargeable battery? You void the warranty. Wanna know what’s even better? Changing the battery means you have to reset everything:
From the manual (pg 26):
Oh, but it gets so much better (pg. 27):
This is every time you change a battery. If my existing pump is anything to go by, this means you have to do this every two weeks or so. So, whether you need to or not, you have to re-prime the pump, re-program the clock, do a setting change, and make sure you have the exact type of battery on hand at all times.
Oh, and how about this for discretion? You cannot turn off the audio bolus sound. According to the manual (pg. 29), “OFF is not an option for this sound setting)” (capitalization in the original). That’s not the only audio that can be turned off:
Fail. Fail. Fail.
As for the Animas Ping – the predecessor for the Vibe – it requires more equipment and materiel to carry, along with an even poorer user interface and feature set. Moreover, the Ping has been criticized for being very loud, especially in quiet environments, which runs afoul of the criteria for discretion.
Reviewers have commented that Animas customer support is non-existent, as is evidently their pre-sales support.[Update: After I wrote this blog I did receive an email in response to my request for more information. I got a couple of brochures from Animas, the same ones that were available on the site, and a DocuSign form to hand over my health insurance information before we could proceed with discussions about the pump. I politely replied that the Animas pump did not meet my needs, and the DocuSign insurance document was cancelled/rescinded, but no other communication was forthcoming.]
According to these results, all manufacturers failed to pass one or more of the listed criteria.
Seriously, what should I do at this point? Return to needles? Surely, that has to be a giant step backwards, right?
Honestly, I didn’t intend to consider it. After all, I haven’t used needles (or insulin pens) in 15 years, but the more I started to think about all the crap that I have to go through in order to cope with living with pump technology – and how far away they’ve gotten from actual treatment methodology – I started to realize that this return to simplicity may be just the thing I’m looking for.
In fact, development with respect to Insulin Therapy (IT) has not stagnated. In fact, several developments in long-acting insulin have happened in the past 15 years. One of the most promising has also been one of the most recent.
In September, 2015, the FDA approved Tresiba and Ryzodeg (insulin degludec injection, 100% and 70/30, respectively) for adults with Type 1 or 2 diabetes. Side effects are in parity with other insulin variants: low blood sugar, allergic reactions, injection site reactions, lipodystrophy, itching, rash, edema and weight gain. In other words, it’s insulin, and acts like insulin.
Degludec differs from other ultra-long acting insulin in that it dissociates slowly after subcutaneous injection, and a binding to albumin contributes to its longevity within the body’s circulation (Kalra, S., 2013). In other words, you get a longer, flatter extended insulin profile, released smoothly without peaks, mimicking ongoing basal transfusions from, say, an insulin pump or normal, working pancreas.
Unlike NPH or glargine insulin analogues, degludec is not constrained by rigid dosage methodologies, requiring fewer doses for basal coverage and allowing for flexibility in dosage frequency. Clinical trials show fewer instances of hypoglycemia and better control of blood glucose (especially when compared to glargine). Concerns about negative effects on cardiovascular risk prevented initial approval by the FDA, but subsequent trials convinced the FDA to move forward with approval in Fall, 2015 under the proviso that long-term longitudinal studies are required to understand the potential risk over time.
Clinical test trials show that degludec insulin result in statistically lower HA1C than other typical long-acting IT methods. No studies were found to determine the efficacy of degludec versus pump therapy, however, but I think that might be something that could be in the works (that would have to be a longitudinal study, and there hasn’t been enough time for that yet).
Overall, I’m encouraged by this. However, in conversing with my doctor, she’s concerned that the newness of the approval means that there is not enough of a track-record to warrant complete trust yet. As a guy who works in technology, I completely understand this!
As a result, I’m forced to consider – for the time being – glargine insulin (yes, the glargine monster joke came back to bite me on the aspart, I know). This requires more of a rigidity in the dosage (that means, I have to do it at exactly the same time each day), which means I’m going to have to figure out how to accommodate a heavy travel schedule that takes me across time-zones.
As you’ve guessed, I have decided to go back to Insulin Therapy. I didn’t actually think that I would ever do that.
This may very well be an unpopular decision. At this point in time, I don’t know of anyone who has decided to revert to non-pump treatment after an initial trial and error phase. At least, I’ve never heard of anyone reverting to non-pump therapy after being on the pump for as long as I have.
Nevertheless, I had assumed (hoped?) that the technology would get better and better, especially as competition forced distinguishing characteristics of pumps to improve over time.
I have been profoundly disappointed by how that has not happened. Instead, manufacturers have competed with each other to make things look prettier and/or more compatible with user expectations (such as mimicking smart-phone technology, as in the Tandem products) and ancillary features (such as Animas’ water-proofness).
The cynic in me is so frustrated with the companies’ lack of responses to basic, fundamental questions that I couldn’t help but wonder if this was intentional.
Tinfoil Hat Time
Diabetic medical equipment manufacturers (I keep this narrowly focused because I have no experience outside of this area) don’t want a cure for diabetes? They don’t want to make peoples’ lives easier or better?
Medtronic, for example, seems to be completely incapable of addressing the life-and-death consequences of pump hacking, and they don’t seem to care.
Animas seems bound and determined to make insulin dosages as complicated and intrusive as possible.
Tandem has substituted effective design for flashy UI, and all at the expense of efficiency and reliability.
To me, the efficacy of an insulin pump is directly correlated to its ability to stay out of my way. None of these options make my life simpler. Every single one of them adds onerous obligations for me to spend more time with my pump, either by design or by consequence, and none of them improve my situation over my older, end-of-life Paradigm option (which, believe me: the best thing I could say for it was that it was tolerable).
What I see is that pump manufacturers have a 4- or 5-year lock-in for their customers. There is no incentive whatsoever to improve the end-user’s experience, and even with the FDA’s meddling in UI (e.g., Tandem’s touch points is partially due to FDA’s nervousness about accidental button presses, ironic given Animas’ horrors with them) the manufacturers need to take on the bulk of the accountability.
I’m particularly disappointed with Medtronic. Animas and Tandem are new companies, and are still in their development phases. Medtronic has been doing this for a long, long time, and I find their unwillingness to answer basic questions about known flaws in their software unforgivable.
Ironically, the only pump worth examining – from an actual product perspective – is the Medtronic Revel, and that’s only because it is the only option close to meeting the remainder of the criteria, even granting wide leeway.
Conclusion, a.k.a. <rant>
I genuinely did not think that I would be returning to traditional Insulin Therapy. However, at this point in time it appears to be the only available option that meets all of my criteria, as well as my airplane litmus test.
Will I go back to pumps? Possibly. Probably. I don’t know. At the very least I won’t have to deal with people making snide remarks about my “pager” for a while.
What I can say, though, is that my faith in the motives of pump manufacturers has diminished to nil, or as close to it as it can come.
It seems that every system I can find is hell-bent on broadcasting – as loudly as possible – to the world that I’m a diabetic. I don’t care that other people are proud of being a diabetic and will create blogs or entire YouTube channels devoted to being a diabetic. I am more than my illness, which is why I try to be as discreet as possible. It’s as if the pump companies think this (taking care of my diabetes) is all I do all day; surely I just screw around with my pump and blood sugar monitoring non-stop, right?
If you’ve read this far, and if you know me, you may be surprised by this sudden outpouring of frustration; I simply don’t talk about it. If you’ve had a conversation with me about diabetes – especially my diabetes – it’s because you’re close enough to me to get me to go against that part of my nature.
Hell, I’m a technologist. I’m an early adopter by nature. I like new gadgets, gizmos, and toys. However, early adopters accept a certain amount of risk. They know that they are in the early stages of new technologies and are willing to accept that risk in exchange for making their lives easier, simpler, or more efficient. But they can abandon early-adopted technologies if their lives are made worse.
This, on the other hand… this borders on pure, unadulterated torture. I can’t believe how much insulin pump manufacturers seem to actively hate their customers. I have a hard time understanding how they – and the FDA – can screw around with people’s lives like this.
Being a diabetic sucks. It is hard enough dealing with injections, equipment, sleeplessness, shakes when your blood sugar is low, the feeling that acid runs through your veins when the blood sugar is high, panicking when your backup of your backup of your backup fails, the inexplicable moments when your blood sugar is 90 at one moment and then 450 half-an-hour later and you haven’t eaten a damn thing.
Being a diabetic sucks, being in meetings and all of a sudden a device attached to your belt screams bloody murder because of a kink in the tubing or the battery dies.
Being a diabetic sucks when some asshole in the TSA has never seen one before (yes, it still happens, especially overseas) and calls over the police to investigate, causing a scene.
Being a diabetic sucks when you can’t pack an overnight bag without your diabetic pump supplies take up more room than your clothing.
Being a diabetic sucks when your doctor tells you that the retinas are pulling away from your eyes, and they can’t do anything about it until things get really bad because the procedure to fix it is more dangerous than simply letting it happen.
Being a diabetic sucks when you get the pins-and-needles feeling in your toes and fingers and it won’t go away for hours upon hours.
Being a diabetic sucks without having to spend every minute of every day worrying that the equipment that is supposed to keep you alive is actually doing its job.
Being a diabetic sucks when you can’t get a straight answer about whether or not a piece of diabetic equipment can be hacked by someone who can kill you.
Even without being melodramatic, the last thing I want is for a piece of equipment to broadcast this part of my life to the world. Until the pump manufacturers get that message – which I seriously doubt they will, now that everybody and her sister seems to want to broadcast their maladies from the highest rooftops – I don’t have high hopes for my own ability to have, or be seen as having, a normal life with pump therapy.
Who knows? Maybe in 2 months I’ll be fed up with the insulin pens and will grudgingly return to the sleepless nights, the “pager” comments, the TSA “preferred treatment,” and the cyborg lifestyle.
At this point, it just doesn’t seem to matter very much. I just want someone who is supposed to be making this suck a little less, actually make it suck a little less, instead of actively trying to make it suck a little more.
For now, it just appears to be too much to ask.
If you want more information about the degludec insulin, for whatever reason, here are some sources (some of which are included above in the text of this article).
Advances in insulin therapy. Kalra, S. http://www.jpma.org.pk/full_article_text.php?article_id=4310
FDA approves two new drug treatments for diabetes mellitus. http://www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/ucm464321.htm
Novo Nordisk Receives FDA Approval for Tresiba® (insulin degludec injection) for Adults with Type 1 and Type 2 Diabetes. http://press.novonordisk-us.com/2015-09-25-Novo-Nordisk-Receives-FDA-Approval-for-Tresiba-insulin-degludec-injection-for-Adults-with-Type-1-and-Type-2-Diabetes
Tresiba (insulin degludec injection) http://www.centerwatch.com/drug-information/fda-approved-drugs/drug/100097/tresiba-insulin-degludec-injection
Advances in Insulin Therapy: A Review of Insulin Degludec. http://www.jfponline.com/index.php?id=22143&tx_ttnews[tt_news]=165854
Type 1 diabetes: insulin degludec. https://www.nice.org.uk/advice/esnm24/chapter/evidence-review
A Review of the Pharmacological Properties of Insulin Degludec and Their Clinical Relevance. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4156782/
Insulin degludec. http://medind.nic.in/jaw/t13/i2/jawt13i2p103.pdf
Clinical use of insulin degludec. http://www.sciencedirect.com/science/article/pii/S0168822715001849
Insulin Degludec and Insulin Degludec/Insulin Aspart: A Review of Their Use in the Management of Diabetes Mellitus. http://www.docguide.com/insulin-degludec-and-insulin-degludec-insulin-aspart-review-their-use-management-diabetes-mellitus?tsid=5
A New Basal Insulin With Lasting Power. http://www.diabetescare.net/authors/samuel-grossman/tresiba-a-new-generation-basal-insulin-with-an-ultra-long-duration-of-action