Rising Creatinine, Myoclonus, and an "8/10" That Doesn't Fit
Mr. R is 68 with metastatic pancreatic adenocarcinoma palliated by a biliary stent; he weighs 58 kg. For three weeks he has taken extended-release morphine 60 mg every 12 hours plus 20 mg every 4 hours as needed, lately using about six rescue doses a day. This afternoon he rates abdominal and mid-back pain 8/10, yet his nurse charts him as "comfortable and sleepy." Over 48 hours he has developed frequent multifocal myoclonus, hyperalgesia to light touch, and drowsiness alternating with agitation. Creatinine rose from 1.4 to 3.1 mg/dL after a contrast study two days ago. He is oriented to person only, pupils are 2 mm and reactive, his abdomen is distended with a firm liver edge, and he has had no bowel movement in five days.
1. The pain score and the nursing note contradict each other. What is your differential, and what data will you gather at the bedside before you change his opioid order?
Read the worked answer
The numeric rating is not trustworthy here. A person who is oriented only to person cannot reliably self-report, and hyperalgesia to light touch suggests the "8/10" is partly provoked by the opioid rather than reflecting untreated nociception.
Leading hypothesis: opioid toxicity from renal accumulation. Morphine-6-glucuronide (analgesia, sedation, respiratory depression) and morphine-3-glucuronide (neuroexcitation: myoclonus, allodynia, hyperalgesia, agitated delirium) both depend on renal clearance. The creatinine rise is the driver, not the dose itself.
Also keep on the differential:
- New or progressing bowel obstruction, perforation, or hepatic capsular distension (firm liver edge, distension, no stool in 5 days).
- Delirium from another cause: hypercalcemia, hyponatremia, infection, urinary retention, fecal impaction, new medications, hypoxia.
- Opioid withdrawal (if he has been under-dosing because of sedation) - less likely with this pattern.
Bedside data to collect: sedation scale score and respiratory rate/pattern; CAM assessment; full medication list for other renally cleared drugs (gabapentin, pregabalin, antibiotics, hydromorphone metabolites); fluid intake, urine output, orthostatic vitals; bladder scan; rectal exam for impaction; abdominal exam for rigidity or tympany; basic metabolic panel, calcium, glucose, BUN/creatinine trend; urinalysis and chest exam; and the patient's own description of the pain's quality and location using simple words. Note also that ascites and poor intake alter volume of distribution and effective dosing.
2. Write the medication plan you would recommend for the next 24 hours, including what you would stop, and justify each change.
Read the worked answer
Stop: extended-release morphine and all morphine rescue doses. Increasing the dose would worsen the toxicity.
Rotate to fentanyl. It is hepatically metabolized to inactive metabolites and is the preferred strong opioid in renal failure. Because he is opioid tolerant, calculate an equianalgesic starting dose, then reduce by roughly 25–50% for incomplete cross-tolerance and titrate upward.
Bridge the gap. A transdermal patch takes 12–24 hours to reach steady state, so start with subcutaneous or IV fentanyl boluses or a continuous subcutaneous infusion for the first day and overlap before relying on the patch. Never start a patch in an opioid-naive patient.
Alternatives worth naming: methadone is renally safe and helps neuropathic and visceral pain, but requires experience and QTc monitoring; if hydromorphone is used instead, reduce the dose because its metabolites also accumulate. Avoid meperidine and codeine entirely.
Treat the neuroexcitation: low-dose clonazepam or midazolam for myoclonus; correct dehydration if consistent with goals; treat any impaction or retention.
Add non-opioid leverage: continue or increase dexamethasone for hepatic capsular distension, consider a nerve block referral, and use haloperidol only for agitation that persists after the cause is addressed.
Monitor: sedation score, respiratory rate, myoclonus, and pain every few hours rather than daily; document the rationale for the rotation so the next clinician does not read "8/10" and reflexively raise the dose.
3. The attending asks whether a celiac plexus neurolysis should be scheduled now. How do you answer?
Read the worked answer
Say yes to a referral, but not as the acute fix.
Why it fits him: celiac plexus neurolysis has good evidence for upper abdominal visceral pain from pancreatic cancer, particularly when systemic opioids are producing toxicity or inadequate relief. If his life expectancy is weeks to months and his pain is largely visceral, earlier referral gives more benefit than waiting until he is too unstable to lie prone.
Why it is not today's problem: he is acutely encephalopathic with an acute kidney injury. The procedure requires 20–30 minutes prone, sedation, and often holding anticoagulants, and it carries risks of orthostatic hypotension, diarrhea, and rare neurologic injury. Stabilizing his analgesia and delirium first is safer and will also clarify whether the residual pain is visceral (block-responsive) or somatic/neuropathic (not reliably helped).
What to say to the attending: "I would place the referral this week as a parallel track, but the next 24 hours need to be about stopping the morphine and controlling the neurotoxicity. If his pain is still refractory on an appropriate fentanyl regimen, the block becomes a priority."
Confirm goals with the patient while he is lucid enough to participate, and document that conversation.