In the auto-mode context “phantom carbs” means telling the pump you’re eating carbs when you actually aren’t, in order to make it lower your BG more aggressively, so that’s a little confusing.
But dealing with exercise lows is a difficulty I have, whether on auto or not. Either way, all the pump can do to raise your BG is to stop giving insulin, and you have to do it sufficiently far ahead because the basal dose you’re getting right now only reaches its full effect later on. In the case of auto mode, you set a “temp target,” since there’s no set basal rate to begin with, and it’s hard-wired to 150. In my case that just didn’t work. I can drop a lot more than that with sustained aerobic exercise, especially in late afternoon/early evening, when I seem to still be generating some endogenous insulin.
I don’t think there’s any trick to make auto any better at dealing with this than manual would be. Basically you can set a temp target/basal—or suspend delivery entirely—well ahead of your exercise time (2 hours?), or eat some carbs before you start exercising, or some combination thereof. Some experimentation is required to get a good sense of how many points you usually drop with exercise and where your BG needs to be at the start in order to avoid going uncomfortably low, but there’s no hard-and-fast rule that applies to everyone (YDMV).
The only other thing I’d add is that, at least in my experience, those exercise lows are subject to a bounce-back effect. Within a half hour of finishing my ride I’ll usually regain 70% of whatever BG drop I’ve had, so I try to be really restrained about “treating” that kind of low right away. But again, YDMV.
How far ahead of your cardio are you reducing your basal? What percentage? If you don’t reduce it an hour or two prior, you lose the effect of basal reduction for exercise. If you bolused within about 4 hours, for food intake, that will complicate things–ie, increase the likelihood of going low from cardio. I reduce boluses to roughly 1/2, if I’m going to getting physical within about 2 hours. For longer times between eating and exercise, I don’t reduce it as much or I’ll spike to much, which I hate.
Thinking ahead to a secondary effect is usually helpful to me but sometimes it moderates my attempt to treat the primary symptom and I end up under-treating. It’s tricky but you make a good point.
Ponder’s sugar surfing philosophy maintains a limited time horizon for treatment. Basically you just need to surf the wave you’re on. Not perfect but another way to look at it.
I’ve tried activating the “temp basal” 2 hours in advance or more, but it doesn’t seem to help. My next strategy will be to exit auto mode and reduce the basal 60%. Kinda crappy that I can’t use auto mode when I exercise. One of the main attractions of the 670G was avoiding hypos.
I am 4 weeks into the 670 g CGM. Initially you’ll find it a rocky road in Auto Mode, but trust me, it gets better and better. This pump is VERY unlike any other pump I have used because over a span of weeks “it gets to know YOU”. You may have crazy high/lows as the Auto Mode micro boluses continue to adjust. Did you know that there is a 670g support group on Facebook? 5,000 participants from all over the world. PATIENCE will be needed and please don’t give up. Your trainer/coach will really help you the first month, then you’re handed over to an Auto Mode coach who will contact you. I talked to mine for 45 excellent, helpful minutes last night. Good luck!
Since I don’t own the 670 I don’t know much about it’s usefullness in auto mode, but my gut tells me I wouldn’t use it. Not just because I wouldn’t trust my life to their sensors, but because my current regimen with a 551 works well for me. I never would use “suspend on low” because MM sensors all tend to show false lows when I’m laying down.
I am not looking to bash any product but I continually hear about the important role that a trainer/coach plays in the 670G.
I am looking for a pump that uses sensor data from a reliable cgm system to automate insulin delivery that does not require anything more than an initial 45 minute session (max) with a pump trainer. Until then, I am glad that there are choices available for people who want to use it but it remains not my choice.
So be careful: buyer beware. I upgraded from Medtronic MiniMed 530G/CGM to 670G/CGM system. Had accuracy issues with previous generation Enlite sensors. Only on 670G system for about a week and am now on sensor #4. Not the “up to 7 days…”, implying that you can expect a week’s worth from each $110 sensor. Brief overlook: sensor one stopped working after 4 days, the second one did not remove the sensor release film with the inserter device and I didn’t notice it until I had it injected and the film was between the sensor and my skin, consequently fell right off. So I couldn’t use that sensor at all. The film didn’t release properly on sensors #3 and 4 either (but I knew to look for it) and had to remove it manually while the sensor was in the serter. (Help desk strongly implied that it was user error, so be prepared for this attitude before you call them so you are emotionally prepared). Third sensor quit after 4 days. Fourth one in today—we will see. FYI: just found out from their help line that they will only replace 2 sensors at a time regardless of your number of problems with the sensors and you must return any faulty/defective sensors. Makes sense but I wasn’t aware of their policies. From my previous CGM research, it sounds like Dexcom users report more compassion towards their customers. Was able to backtrack and dumpster dive to retrieve all but one sensor so that’s all they will replace.
Someone here who also gets only ~4 days out of her Guardians recently reported that Mt told her that individual body chemistry can affect on sensor life. In my case I regularly got 14 days and only had to change when battery level got too low. So it definitely varies widely. But it’s a pretty common complaint from the groups I’ve been on.
Hi. I’ve used the Animas for 10 years as the Ping and Vibe. I never had pump problems other than them not being truly “waterproof “. They are now out of business. Just curious what problem you had with them.
Hi Talisker,
I was just curious how the 670G has been working since this post. I have been thinking about upgrading but have had issues with the Medtronic CGM’s in the past. So I know use Dexcom and my 530 pump.
Thanks.
After 7 frustrating months with auto mode which led to a decision to go back to Manual Mode…AM is working. It’s working for the most part because I have gone to a very low carb diet.
