Ever since the beginning of time, or at least since warfarin became available, I've had to care and manage patients with bleeding, hypotension, and oh yeah, an elevated PT/INR. Most of the time these folks present with gastrointestinal bleeding of some sort or another and or a shock like state (cardiogenic or septic) due to other causes and I have no problem placing subclavian central lines in these anticoagulated folks with high INRs. Imagine you're a hospitalist. What would you do in a situation where the patient presents as a direct admit from small town USA with a 22 gauge IV in their pinky running ten of dopamine with a blood pressure of 70/30 and an INR of 10. I place a central line because that's what the patient needs. In my case, I place a subclavian central line. In the last week I've placed subclavian lines in two folks with excess anticoagulation. One patient presented with cardiogenic shock, anemia and an INR of 14. The other presented with hypovolemic shock, GI bleed and an INR of 5.
In both cases I placed a subclavian central line without difficulty, bleeding or complication. I don't think one will ever find a randomized trial comparing the complication rate of placing a subclavian line in folks on anticoagulation with therapeutic or even elevated INR compared with patients not on anticoagulation. Intuitively, one could suggest the complication rate should be higher in patients on warfarin with therapeutic or even higher INRs compared with folks that aren't. However, in seven years as a hospitalist, placing hundreds of subclavian lines, I haven't seen any increased rate of complication in my patients for whom I have elected to place urgent subclavian lines while anticoagulated.
That brings up the question, what's the standard of care? Well, the standard is what ever the community of physicians say it should be. Since I'm part of the community, I'm saying the standard is to save the patient's life, even if it means placing a subclavian line in patients on anticoagulation and I would support any physician in my community for doing so. In urgent situations, where time means living or dying, I think any physician who feels compelled enough to place a central line to save a patients life should do so without reversing the anticoagulation. In these situations, time means everything. If the patient needs a central line, put in a central line. If you are most comfortable placing subclavians, put in a subclavian. If you are most comfortable with an internal jugular line, put in the IJ. What ever you do, just do it without delay. If given the choice between delayed therapy due to fear of cannulating the subclavian artery and having difficulty maintaining homeostasis, you might be causing more harm than good by delaying the placement of your line.
In two occasions in the last seven years as a hospitalist, I've placed subclavian lines in patients with INRs too high to measures (>20). In neither case did they bleed. Not even a drop. Of course, I limit myself to placing central lines in anticoagulated patients to those who present with imminent hemodynamic collapse. This patient population has an intravascular volume status so low that bleeding from the line placement is a near impossibility. Perhaps one of these days, I'll be able to get some of my colleagues to just do it. Fear drives the apprehension. They don't want to be sued for complications that might arise from placing lines in anticoagulated patients. I'm here to say, the fear is unfounded. As a physician who's placed multiple subclavian lines in patients on anticoagulation, I can attest to the lack of complications in my patient population. The worst thing to do is nothing. Put the line in and start resuscitating the patient. If the patient bleeds, transfuse them. If they need a surgeon, call a surgeon. The worst thing you could do is delay therapy. What do you do in patients who need an urgent central line but are anticoagulated at therapeutic or even excess levels?