Celiac Plexus Block on Apixaban With Thrombocytopenia
72-year-old woman with unresectable pancreatic adenocarcinoma and severe epigastric pain radiating to the back. She takes apixaban 5 mg twice daily for atrial fibrillation (CHA₂DS₂-VASc 5) and has a creatinine clearance of 42 mL/min. Platelets 88,000/µL, INR 1.1. Her oncologist requests a celiac plexus neurolytic block. She is on oxycodone 20 mg every 6 hours with partial relief and has had two falls in the past month. The interventional fellow asks whether to proceed tomorrow and whether to stop apixaban.
1. How would you weigh the bleeding risk against the thrombotic risk, and what would you do about apixaban?
Read the worked answer
This is a high-stakes trade-off. A celiac plexus neurolytic block is a deep, high-risk procedure near the aorta and celiac artery, so ASRA-style guidance would have apixaban held for about 72 hours before the procedure, assuming stable renal function; with CrCl 42 mL/min, the drug's half-life may be prolonged, so I would confirm the last dose and consider a longer hold or checking a calibrated anti-Xa level if there is uncertainty. I would not bridge with heparin, because bridging increases bleeding without proven benefit for DOACs. Her CHA₂DS₂-VASc of 5 means a real stroke risk if the hold is prolonged, so I would coordinate with cardiology and the patient's oncologist, document the shared decision, and restart apixaban 24 hours after the procedure once hemostasis is confirmed. Platelets of 88,000/µL are above the usual 50,000 threshold for most deep plexus blocks, but I would check a recent trend and avoid the procedure if there is active bleeding or a falling count.
2. Would you proceed with the neurolytic block tomorrow? Explain your reasoning and alternatives.
Read the worked answer
Not automatically tomorrow. The benefit is plausible because she has pancreatic pain with partial opioid response, but the combination of apixaban, thrombocytopenia, falls, and recent opioid escalation means the risk-benefit ratio is not clearly favorable. I would first optimize analgesia with an opioid rotation, scheduled acetaminophen, and possibly low-dose ketamine or a gabapentinoid, and assess whether the pain is truly somatic/visceral pancreatic pain. If the block is still desired, I would schedule it after a documented 72-hour apixaban hold, correct any coagulopathy, use CT or EUS guidance, and have blood products available. An EUS-guided celiac plexus block by an experienced gastroenterologist may be preferable because it avoids a percutaneous route and may reduce bleeding risk. A splanchnic nerve block or continuous celiac plexus catheter is another option. If she cannot safely hold anticoagulation, I would not perform a neurolytic block and would focus on pharmacologic and palliative strategies.
3. What would you do if the patient refuses to hold apixaban because she fears a stroke?
Read the worked answer
I would explore her concerns and explain that the risk of stroke from a 72-hour hold is low but not zero, while the risk of major bleeding from a deep neurolytic block while anticoagulated is substantial and potentially catastrophic. I would offer alternatives that do not require stopping apixaban, such as optimizing oral and transdermal analgesia, EUS-guided block with a gastroenterology team that may have a different protocol, or a temporary celiac plexus catheter if the risk is judged acceptable. I would involve cardiology to quantify her stroke risk and document the discussion. If she still declines, I would respect her autonomy and not proceed. The key is not to coerce her or to proceed silently while she remains anticoagulated.