Getting visible blood on the CGM sensor tube alters accuracy?

Hello all,
I have recently switched, after twenty years of using shots, to a Medtronics pump and CGM (Revel). Its been a week now, the pump is great and CGM was working fine. But yesterday I had to change the sensor for the first time. I am a skinny guy (aprox 9% body fat) so finding a place for the sensor is challenging. The best spot for the first sensor was on the left side of the waist a couple of inches above of the waist line.
I put the second sensor near that first spot, but then I noticed the sensor canula filled up with blood. According to the sensor manual, as long as blood stops coming out, then things should work.
But what I am noticing is that the CGM BG values are all over the place (ie, sensor predicting a low when my BG was 177 and going up). I am using the same calibrating protocol I used for the first sensor.

The question is: If blood goes up the canula then does that mean the sensor accuracy will be affected, and I should discard the sensor and apply a new one? The first sensor did not fill with blood and work well. Now I have tried twice to apply a new sensor and both got blood and were not accurate in their readings.

I will consult my pump trainer but community feedback is always welcome.

Bleeding is very common, especially upon first insertion. It's quite a piece of hardware to get in there. The catheter needle is quite a whopper, and personally I've never had much luck with the auto inserters. The silhouette inserter used to make me gush blood. The articulated needles used to dive when given a shove from the device, and so I started manually inserting them.

I've tried manually inserting the sensors too, but the needle seems a little too girthy for that and I just cannot heft them in. So, back to the inserter I went.

I'm not as skinny as you. Not huge or anything, but I've got some real estate to work with here, and I still regularly get bleeding happening. It doesn't seem to really effect anything that I've noticed.
There's no exchange of fluids from sensor to site to anger the wound past the point of making it, and so as long as you're not in discomfort, and the bleeding subsides, the worst problem you have is that it can effect the adhesive and shorten the life of the sensor by coming un-adhered and causing a premature jettisoning of the sensor.

If it makes you feel any better I had a sales rep install an initial one for me to do the three day scan thing they like to offer sometimes, and he had me bleed too. He said it was very common and didn't effect anything.

As far as it effecting accuracy, as I said, as long as it stops bleeding quickly and you're not feeling discomfort, it seems somehow (even though it seems wrong) irrelevant to function.

Slightly off topic...

I notice you're on a bike there too. I'm just getting into biking. Do you find the lag on the CGM to still be of use while riding?

What I am noticing is that the sensor seems to be "over stimulated": It continuously trends downwards event though my BG could be 200 mg/dl. Maybe the direct contact of blood with the cannula causes that.
Regarding the bike, I only used it twice. It was useful to know the trends, but I would be have it as a secondary way of knowing where your BG is. I did increase the low threshold for the alert (from 80 to 100) to have a bit leeway if I was trending downwards.

The sensor is made out of a piece of tubing but one side is cut away after it is filled with a gel and the three sensor anodes are inserted. I have been using the MM/RT system (CGMS) for over 5 years and many of my sensors bleed when inserted. It can take several months for you to learn how to get the most out of the CGMS. Here is a couple of photos showing the sensor construction.

The getting used to it comment is very true, Gus.

I hope you'll stick with it and get as much out of yours as I have mine. I went through all the nonsense with training and so forth, and ended up being more confused than when I started.

The reps do not stress how important calibration is and how pointless it is to even do it when your glucose isn't stable.
If you're experiencing false lows that are continually decreasing regardless of your glucose, it can mean a few things:

1: A very young and recently inserted sensor that is still bedding itself in. It varies from sensor to sensor how long this takes, but after having had these put in by the reps and told to start it straight up, I can look back, and while it's understandable due to their time constraints, also see that this was a very stupid thing to suggest or recommend.
The sensors can take 12-24 hours to stabilize before becoming useful. It's best to put one in at bed time, go to sleep and then try to connect in the morning.
The isig numbers generated can fluctuate madly otherwise. You may even have one where it seems stable and then dies. Leave it another 6 hours and try again. It may become as solid as a rock from that point on. They're temperamental.

2: The sensor is old and about to die. You'll find the isig numbers very slowly falling until you can be completely out of synch. This can also show madly fluctuating highs and lows that display as a peppered line on the bar graph on your revel.

3: Sensor site is too close to your infusion set. This one is fascinating. Once I was fully comfortable with my glucose monitor I had this happen and was able to experiment with it. It also showed me how long my insulin was lasting in my body, because even after I moved my infusion set, I could still chart this incredible false low, that very very slowly wore off in roughly 4.5 hours. (as the insulin surrounding the glucose sensor absorbed and went away).

As a slight side note, I've used sensors until they die, and had their lifespans vary enormously. I've had some last 6 days, and twice, I've had my longest last a 23 day stretch. The one I have in mind for this story was reliable and stable right up until that last day.
I couldn't tell you how or why they vary so much, but I do remember that the long lasting set in question was a bleeder and I had to keep cleaning dried blood off the area as I had to recharge and then reattach the transmitter so many times in it's life.
It did make me question if perhaps a little inflammation at the site might be actually helping it obtain a decent signal for so long.

Any help?

Thanks. This helps indeed. The sensor finally stabilized after around 20hs , so next time I will insert the new one the night before I switch. Thanks!

Thanks John, this is very helpful.

Try the back of you arm, mid way between shoulder and elbow. This is the most consistent reading location for sensors according to the clinical studies. Try an angle the sensor a bit more to get the tip at the right skin depth.

As noted the sensor is not affected much by the blood as long and it is not where blood is flowing.

I've been using the same arm in about a 3"x3" area for 5 years now. Can't wait for the new MM sensor/pump to come out. The new sensor will be good for 6 days and is far more accurate than the 5 year old Revel technology.

My readings are usually fine regardless of how much blood there is. The Opsite Flexifix tape will hold quite a bit of blood/ sweat and as, long as the data is ok, I'll keep using it.

Re the off readings, I don't get that many of them but usually will just ride out the time period until the next calibration and try to get my BG to somewhere near where it's at. If my BG is 177, I would usually fix it one way or another and doing another cal generally doesn't help that if I'm moving my BG. If I have a stable number that's ok (120 or less maybe?), I might consider firing another calibration to see if it will knock some sense into the reading. I just had one that I did that to yesterday and it didn't work. I wasn't sure what was up with the thing.

Here is a tip for bedding the sensor: If I am changing the next morning, I will insert just the sensor the night before, leaving the current sensor working for overnight protection. Then I get up, disconnect the old sensor and charge the transmitter. Then I attach the transmitter and shower.

Then I link the sensor and it usually asks for a reading within 5-10 minutes, but not always.

Alternatively, you can charge the transmitter, then insert the sensor. I will wait 4-5 hours before linking in this scenario.

And I wear each sensor for 7 days. After that, the transmitter requires charging.

Thanks all for the responses and great insights.
Finally, what I am doing to put the sensors on my skinny body is:
- I use a less aggressive angle, around 35 degrees. (lesson learnt the hard way here, I used around 50 degrees when trying to put a new sensor about two inches to the right of my abs, profuse bleeding started, removed the sensor, and had to use four paper napkins until the bleeding stopped. And it hurt for a bunch of days. My first self-inflicted stabbing wound, yey!
- I am using the right side of my body for now exclusively above the waist line. I couldn't dare try to put this thing on my arm.

Things are working fine now, I love this CGM thing.


Glad to hear it, Gus. I know that acidrock guy there is on the lean side of the equation too :)
I bet he'd have some site recommendations for you.

Honestly, I'm about the largest I've been in my whole life and I'm not sure I could do an arm one either. I'm basically certain that the next gen of these things are going to be considerably smaller.
Lets face it, you could murder a bear with one of these needles.