Great discussion, and I appreciate the input very much. Here's how it went down...
I discussed in plain language the possibility of cardiac arrest / death during the operation and subsequently on the floor. They understood and realized this was a high-risk situation.
My first choice was a carefully titrated epidural, but for obvious reasons, this was not practical. I opted instead for a light GA with an LMA. I placed a pre-induction a-line (thank god) and induced using lidocaine, propofol and a smidge of fentanyl (12.5 mcg). I chose phenylephrine as my pressor because I reasoned it would adequately counteract the vasodilation of the propofol and with any luck, I'd keep her at baseline. I also use it a lot more on a daily basis. I did have epi (10 mcg/ml) immediately available. I gave the propofol in 10 mg boluses and waited knowing she had a slow circ time. At 40 mg of prop, she began to drop -- like a rock. I began pushing phenylephrine but with no effect. I emptied 1500 mcg and watched the pressure fall -- to ZERO. PEA arrest. As it became clear she was going to arrest I called for backup and when her a-line flattened out (no change in rhythm) we began compressions. By this time I had pushed a 50 mcg of epi and after about 30 seconds of vigorous compressions, we had a perfusing rhythm. We thought about it and decided to proceed with the case knowing her PA pressures and her heart were no going to improve, especially after her PEA arrest. We relaxed her, changed the LMA to an ETT, lined her up and did the case. As the case went on she developed significant ST depressions which we really couldn't do much about given she was now on a levophed gtt.
We took her upstairs to the unit and the family withdrew on her the next day!!! WTF??!!
I think what happened is this: unbeknownst to me, my CRNA had slipped her 1 mg of midazolam in the HOLDING area. Holy **** was I pissed when I found out that. But because I'm the MD, the responsibility lays with me, not her -- and I wouldn't have that any other way. It was my fault for not controlling her more effectively. She undoubtedly hypoventilated from the versed. She also likely hypoventilated during the induction: given we were pushing the prop slowly and she was spontaneously breathing the whole time, we never effectively ventilated her. This caused her PaCO2 to increase and her PVR to follow suit, effectively stopping forward flow. This exacerbated the hypotension from the propofol and mitigated the effect of the pressors -- and presto - PEA!!
What really ended up helping her to maintain a perfusing rhythm (in addition to the compressions) was effective ventilation.
Lessons learned (for the residents reading this):
1) Respect pulmonary HTN. Hypercarbia, acidosis, hypotension and hypoxia can kill patients with severe pulmonary HTN.
2) Midazolam is not a benign drug that EVERYONE gets prior to surgery. Evaluate on a case-by-case basis. See above.
3) Control your CRNAs. Be a jerk if necessary. This can be tough to do, but it can be a matter of life and death.
4) Rule #1: The patient is the one with the disease (Read "House of God" if you don't know what I'm talking about)
5) Special thanks to the guys for discussion FI blocks. I've never done one, but they don't sound too tough with a little reading and watching video.
Cheers!