Medical question - how can an "IV push" cause gangrene?

I heard about this case on NPR today: http://mobile.washingtonpost.com/detail.jsp?key=310028&rc=to&p=1&all=1#_1
What I’m curious about is how miss-applying this drug (inserting into artery instead of vein) causes gangrene. Any ideas?

Phenergan is a vesicant and can cause tissue blistering and even necrosis if it seeps into the surrounding tissue. My guess is that there is a higher risk of seepage occurring in arteries due to higher pressure within the arterial walls.

Artery walls contain smooth muscle fibers, which contract under a variety of stimuli. Small and medium sized arteries can go into spasm, where the flow of blood is completely cut off despite a normal blood pressure. Veins cannot spasm in this manner, although they can be irritated chemically, become inflamed, and clot off. A clot in a vein is unlikely to threaten a limb like spasm and/or clot in an artery can (although venous clots have their own risks).

Arterial vasospasm can result if the phenergan is injected directly into an artery (uncommon), but it can also happen if the concentrated medication is ‘pushed’ into a vein under high pressure, leaks out of the vein into the surrounding tissue (extravasation), and bathes the outside of a nearby artery. If the medication is in the extracellular space, it takes a long time to be washed out, so the effect on an artery can last for many hours or even days - thus gangrene.

Thanks for the explanation. I was aware that IV chemotherapy drugs like Nitrogen Mustard needed to be administered carefully, due to the possibility of severe tissue damage, but it’s really surprising that an “anti-nausea” drug could have such serious side-effects.

Promethazine can have a direct caustic effect but the mechanism of causing gangrene in this case is unclear. I doubt the IV was place intra-arterially, although that has been advanced as an explanation (probably since the gangrene was distal to the IV site). An artery has more pressure than a drip IV and so the drug would not drip into an artery; the IV would back up with blood. The explanation is probably not as simple as arterial spasm, and seems to be more of a rare idiosyncratic reaction.