Horrified: possible accidental arterial infusion

  1. I had a patient yesterday who was hypoglycemic. Previously, I'd placed her IV in that diagonal vein in the AC that I've used 100x... Or so I thought. She had an episode of hypoglycemia and had to get D25%. I started slow pushing it, but she complained of warmth and feeling flushed, and her arm was looking flushed. I had probably given 2 cc at that point. So I diluted it in 100 cc NS and hung the infusion over 20 min. She had another episode of her distal arm getting flushed and mottled, though it remained warm with sensation, a strong pulse and no edema. I slowed the infusion more, her arm started to go back to normal. The whole time, the line flushed and drew without pain or complications. Later, the oncoming nurse said he thought the line was pulsatile!!! I totally trust him, but I never noticed a pulse, there was no back flow without me pulling back, and the blood I drew was dark red. I even let it free-flow for a second and it didn't spurt. The tech who d/c'd it said it didn't bleed when taken out. Does it sound like it was arterial?! I am so upset to think I might have put DEXTROSE of all things into an arterial line!!! I have called the floor she's admitted to 3times and the nurses say her arm is fine and she's not complaining. Last time I checked on her was 16 hrs after this whole mess. How soon would tissue damage show itself though!? I can't think about anything else until I know she's going to be OK!!!
  2. 23 Comments

  3. by   Been there,done that
    Most pumps will not obtain enough pressure to deliver into an artery.
    The issue is ...the arm was mottled and you continued the infusion. The purpose of giving dextrose is to rapidly raise the blood sugar, slowing the rate of infusion was not an option.
  4. by   Southgoing Zax
    Well, I stand by my decision to continue the infusion if it was in fact a venous line. It flushed and drew, and was showing gradual improvement. Continuing the infusion was the lesser of two evils- my patient had no neuro issues yet, but given her glucose level, she had to be darn close! Slowing the infusion simultaneously acheived 2 goals: raising her BG and maintaining the line. I should mention that despite the low reading, pt. was entirely alert and neurologically intact. Long story about why she couldn't take PO, and it's too specific to post. The MD's were also in agreement with this plan. What I really want to know is the timeline of tissue damage after such an event, especially if it was arterial. The pump gave me no trouble BTW.
  5. by   Been there,done that
    I remain confused. Why are you concerned about tissue damage, if the glucose entered the circulatory system via an artery?

    BTW , having an MD in agreement with the plan "Don't impress me much"
  6. by   Double-Helix
    If the dextrose had infiltrated and caused tissue damage, you would know relatively quickly. Certainly signs would not arise 16+ hours after the med was given.

    The risk with giving medication through an arterial line is that arteries are very sensitive to medications and can easily spasm. If the artery clamps down, it can cause peripheral tissue infarction from lack of blood flow. Severe tissue damage will be evident almost immediately. (When cells die, and are no longer receiving oxygen, you can tell.) The danger with dextrose is that it's more likely to cause infection. If an artery spasms, the peripheral tissues aren't getting blood flow, so the area would become cool and pale, not flushed. A flushed extremity sounds more like a venous spasm- where the blood pools in the extremity but cannot return- causing a flushed and warm extremity. Venous insufficiency could cause mottling as well.

    Of course, it's impossible to say now whether the line was actually arterial. But just seeing dark blood and making sure the line doesn't spurt are not for sure signs that it's venous either, especially in patients who might have compromised circulation to begin with. The most accurate way to determine where the line is located is with a blood gas or a color doppler ultrasound. What was running through the line before this? What was the rate? Did the patient have any complaints?

    Like the first poster said, continuing the infusion at the dilution/rate that you were running it doesn't really make sense, in this circumstance. The purpose of D25 is to raise the blood sugar when it is dangerously low. D25 dilute in 100 cc of saline dilutes the dextrose to about 2%. Then, running it over half an hour means that the patient is getting a tiny amount of dextrose each minute- not enough to raise the blood sugar to an appropriate level if it really was so low that D25 was required. At that point, if you're trying to give a medication in am emergency and you're concerned about the patency of the line, you're better off giving something sublingual while you obtain new access.
  7. by   Southgoing Zax
    Ah, that makes sense. It definitely was never cool or pale. The line was previously saline-locked and was able to be flushed without pain, discoloration, resistance, flushing/blanching, etc. believe it or not, we don't stock SL glucose(!) but that's an idea. Lesson learned. I wonder if a venous spasm could cause the irritated vessel to appear pulsatile for a time? I have never seen such a thing. Does it have to do with concentration of the infusion or speed of push? Does heat (vasodilation) help? So, tissue damage from arterial spasm is quickly evident then, right? Anyone have an example they'd be willing to share?
  8. by   IVRUS
    Anything above D10 is hyperosmolar and in this case, has an osmolarity above 1000. D25 is a vesicant and can cause severe tissue necrosis and can cause the patient to lose their limb. Whenever a vesicant is administered, you as the medical professional should be ascertaining a brisk blood return with every 2-3 mls of your IV Push and ensuring that the patient has no complaints, or that you are not noticing any adverse affects with your administration. And... If you question, even with a blood return that you are not in the vein, you should NEVER "just wing it" and continue with the administration, whether that be IVP, or diluting it and giving it over another time frame... (You also did not get the Blood Glucose levels up where you needed them when you diluted it.) Hopefully, this is a learning experience for you.
    Also remember that when a vesicant goes into the SQ tissues, the initial tissue sloughing and subsequent necrosis may not be evident for 3-4 weeks. Keep monitoring the area.
  9. by   Southgoing Zax
    Is it possible that there was infiltration despite NO EDEMA, disappearance of all symptoms and no pain/swelling/edema/ discoloration or any problems, now nearly 24 hours after? I have seen infiltrates and extrav's before, and never such a delay in symptom onset. I see what you are saying about late onset necrosis, but is this with or without persistent symptoms in the hours following the infusion?
  10. by   iluvivt
    IVRUS is correct in that tissue damage from a vesicant may not be apparent for several weeks. An extravastion and infiltration are not terms that can be used interchangebly. It does sound like you had an arterial placement based upon your explanation of the tissue flushing and blanching. If you were near the basilic vein it is not that difficult to hit the artery instead . We have a little saying here about PIVs it's "when in doubt take it out" and I would have definately questioned it based upon your excellent observations. You should not base your decision to leave the line in place on the fact that you were able to easily instill and the patient was not having any pain b/c that does not exclude an arterial placement. Many times they do not have pain and may not even pulsate easily. I know of several cases in which an arterial line was inadvertently used as a venous line for several days..and yes the pump continued to pump the IVF. I noticed on the most recent one I happened to catch..just such a case...but I could tell b/c when I checked for a blood return..it was just way too brisk.
    There is slight possibility that the patient may have had a small arteriovenous fistula. I remember one time I accessed the accessory cephalic vein on this patient....it was very large,soft,and on the surface..I felt it..it was a vein..but after I accessed it ...it was behaving just like an artery..very strange.....I took it out b/c I suspected and AV fistula...does not happen very often..but I have started thousands upon thousands of IVs.
    Contrary to popular belief the ACF veins are not desirable for routine IV therapy for a variety or reasons...inadvertant arterial cannulation,very positional, at an area of flexion,should be preserved as blood drawing veins, IVs should be started optimally from the hand up (preserves available proximal veins for use). I know the ED likes to use them and I deal with it....and they are usually easy to hit,but if the patient beomes an admit....we just have to change them anyway b/c the pump beeping drives the patient crazy. I do use them for power injection if I have to..but then I have them d/ced after that and resite it in a better location.
    Last edit by iluvivt on Jul 3, '12 : Reason: spelling
  11. by   brainkandy87
    Well, I don't know what floor you work on, but hopefully your ER has a portable ultrasound for inserting IV's (your cath lab def has one). That's the best way for confirming placement. I'm called a lot to the floors to insert lines with US and have seen some funky anatomy and felt some strange veins. Once, I put an IV into an artery that felt 100% like a vein and it was non-pulsatile, painless when flushed, no apparent skin issues, and the blood draw appeared ALMOST venous. It could've passed for venous. I just suspected something wasn't right.

    So I pulled out my US and started looking. Sure enough, it was in her artery. Her AC anatomy was just different than most. Her artery was sitting directly on top of her basilic vein and very superficial. Learn to use US. It's a very valuable resource. If you can't find a vein, you won't have to question what you're poking. Also, if you're ever unsure if it's arterial and you don't have access to US to visualize it, pull it. Poking the patient again is much preferred to the adverse effects of an accidental arterial infusion.
  12. by   jadelpn
    If it were an existing SL, especially in the AC, it can and does infilltrate especially if the patient is bending arm. The moment that I saw that there was any irritation, discoloration, or met with any resistance,, would be when I would call a co-worker to attempt another IV in an alternate location, while I monitored the patient. Or your charge nurse for assistance. VERY important to make sure that you have no IV's needing rotating, or flushing with your assessments. The time to find out you have a bad IV is not when you are needing one stat. I understand that timing is of essence when someone is hypoglycemic and you need to get the dextrose in, but you need to err on the side of caution and what you are describing sounds to me like an infiltrate. Also be mindful of the IV gauge. If you are pushing dextrose through a 22, that is not easy, and can cause issue. IV's that are going to stay in are not ideal in the AC. If you have to get one in quickly for pushes and the like, that is one thing. If you have one that is staying in, then look for other sites than the AC.
  13. by   iluvivt
    Agree ,it's not uncommon to see the artery on top of the vein or to see an extra artery with US. Yes... you can use bedside US as one way to check your placement but you did a good assessment and had all the clues you needed to make the best decision and pull the line. I am not a great fan of US assisted PIVs though I do perform the procedure when needed. In a recent studies they have been found to have a high failure rate when compared to PIVs started in a traditional manner,with failure rates as high as 50% in 24 hrs. I have found this to be true. I have also found that the nurses have a much more difficult time assessing the sites and the infiltrations I see are much larger and caught later in the process of infiltrating. They do have a place in certain circumstances when specific and prudent guidelines are followed especially as a bridge line until a more suitable VAD can be placed if venous access still needed.
  14. by   Southgoing Zax
    So...just to clarify one more time. I understand that tissue damage from an extravasation or and infiltrate may not show up for weeks, but what about tissue damage related to an accidental arterial infusion? Given that this is not the same as an extravasation or infiltration, since infusions were entering the circulatory system and NOT the surrounding tissue (albeit in the wrong place) is there any reason to suspect that late-onset tissue damage will be a problem?