Sedation and analgesia III: Non-opioid analgesics

The third in a multipart series on the basics of ICU analgesia and sedation. Start with Part I: Principles to learn about our three fundamental rules, then Part II: Opioids to learn about “the good stuff.”

You can manage almost any amount of pain using higher and higher doses of opioids. However, you shouldn’t.

Opioids have predictable adverse effects. They cause sedation, limiting mobility and promoting delirium. They suppress respiratory drive, increasing time on the ventilator. They slow gut motility, causing constipation and increased gastric residuals. All of this slows the progress of healing, increases ICU time, and prolongs post-op stays. In short, the less opioid you can use while still effectively managing pain, the better.

Thus, the goal is not a monotonal opiate hum, but a many-layed chord of analgesia. This is multi-modal pain management. Many ICU patients have mild to moderate pain only, and this is easily managed, but for those with significant ongoing pain, such as post-traumatic or post-operative pain, an intelligent regimen should be used that combines short- and long-acting opioids, IV and PO agents, and complementary non-opioids. Let’s discuss the latter.


You would be forgiven for thinking that good old Tylenol doesn’t do much for pain. And small doses given PRN are indeed only suitable for very mild pain, such as an annoying headache.

For real pain, a different approach is needed. Acetaminophen should be high dose and given around the clock. For patients without risk factors, the maximum dose is 4 grams per 24 hour period. Make use of it: give 1,000 mg (1 gram) q6h, scheduled. This will build up a significant baseline of analgesia. For elderly patients or those with somewhat uncertain liver function, you might limit it to 3,000 mg daily. For others, of course, such as those with real liver disease or active shock compromising hepatic function, you may need to skip the acetaminophen altogether.

IV acetaminophen is available. By and large, it is not “better” than the enteral version—although its more-rapid delivery can make it seem that way—and dosing is equipotent. It is simply an option for patients who cannot take enteral meds. It is also far more expensive than tablets, and using it in a scheduled fashion can cause a major drain on your budget. While useful for the immediate post-op period until patients can take pills, don’t leave it ordered for days, lest your administrators slash your tires.

Lidocaine patches

Topical lidocaine patches (common trade name of Lidoderm) are an interesting option. Sometimes, they do little. Sometimes they seem very effective. They seem best for fairly localized, superficial pain, such as incisional pain or rib fractures, and are usually changed daily (such as 12 hours on, 12 hours off).

They have essentially no side effects, and so are never a bad thing to try, although they do have some associated cost.

Non-steroidal anti-inflammatories

NSAIDs such as ibuprofen, or the IV/IM ketorolac (Toradol), are often very effective analgesics—especially when using the above strategy of giving real doses on a scheduled basis. Unfortunately, they are often contraindicated in the ICU, for one of the following reasons:

  1. They reduce GFR, in patients with renal insufficiency
  2. They impair platelet function, in patients at risk for hemorrhage
  3. They can compromise bone healing, in patients with long bone fractures
  4. They can cause GI bleeding in anyone

Since that list includes most of our patients, these drugs often don’t play a role until patients are entering the sub-acute phase of their stay. At that point, however, as you try to taper off the opioids, don’t forget them.

Regional or neuraxial anesthesia

A growing sub-specialty of anesthesia, as well as a niche activity in other fields like emergency medicine, regional anesthesia involves the local application of nerve blocks by subcutaneous infiltration of agents like lidocaine and bupivicaine. Skilled pain specialists can completely block pain to almost any area of the body. Better still, epidural catheters can allow continuous analgesia drips directly to the aggrieved spinal levels, even with PCA functionality.

Although epidurals may still cause hypotension and other complications, they generally yield very dense analgesia with far fewer systemic effects than the blunt use of IV drugs. They are an excellent tool for post-op care after selected procedures and for painful rib fractures that compromise breathing. If an epidural service is available to you, make use of them. More on this somewhat specialized topic another time.


In recent years, ketamine has become the hottest thing to wear a syringe in the emergency department and the ambulance, and is starting to find its way into the ICU as well. It is a non-opioid agent which instead causes dissociation: it leaves the higher functions of the brain intact, yet disconnected from bodily sensations. Thus, there is not only little to no respiratory depression, and it is hemodynamically very stable, usually causing not suppression but stimulation of blood pressure and heart rate (although in severe shock, as with any sedative, it can still cause decompensation).

At low doses, it is a potent analgesic, effective even in patients resistant or tolerant to opioids. Slowly push .1–.3 mg/kg (usually 10–20 mg in adults) every 10 or 20 minutes, or better yet, infuse it in a minibag, since a fast push tends to cause more side effects.

At higher doses of >1 mg/kg, it causes complete dissociation, essentially a state of general anesthesia with no connection between mind and body. Anything from surgery to intubation can be performed in this state.

The trouble is that in between these dose ranges, a state of partial dissociation can occur, where reality is altered, and the mind-body connection is patchy but not absent. At the lower end, and in the right population, patients can find this enjoyable; indeed, ketamine is a recreational hallucinogen and club drug. At the higher end or in susceptible psyches, this results in a “bad trip,” and can cause severe psychological distress and agitation.

You can easily avoid this by either sticking to low doses, or jumping straight to full dissociation with a high dose. In the latter case, however, the patient will still pass through the intermediate dose range as the drug gradually clears, causing the phenomenon known as an “emergence reaction.” Distress can usually be minimized by good mental preparation, creating a pleasant environment and explaining the oncoming trip to the patient. Should unpleasant emergence occur, a small dose of benzodiazepine is effective.

Ketamine in the ICU is still young, and hence its use can be limited by staff inexperience—such as protocols against it being pushed by nursing, or infusion pumps without settings for ketamine infusions. Theoretically, it can be a useful adjunct in unstable patients who won’t tolerate opioids, or in opioid-tolerant patients who don’t respond to their use at all. Unfortunately, there is no real evidence for its long-term use in the critically ill, so it is uncommon to leave patients on this drug for prolonged periods.

Miscellaneous drugs

Various other drugs sometimes make an appearance.

Agents like gabapentin are a good choice for true neuropathic pain, but sometimes will also be thrown into the mix for other varieties. This is of questionable effectiveness.

Tramadol can be used as a sort of compromise between opioid- and non-opioid analgesia. It may offer the worst of both worlds, however, and offers significant risks in ill patients. I tend to avoid it.

Finally, the long-acting opioid methadone sometimes has a valuable role. Most often, this is in patients already using methadone at home, who are best left on it to avoid withdrawal. (Note that by hospital policy, you may need to personally confirm their dose with their methadone clinic before ordering.) However, patients receiving opioids for prolonged periods—who are unable to be weaned—may be transitioned to methadone instead of left on drugs like oxycodone with greater abuse potential and a worse safety profile. This situation is often seen in burn patients.

Next time in Part IV, sedative options for the critically ill.

2 Replies to “Sedation and analgesia III: Non-opioid analgesics”

Leave a Reply

Your email address will not be published. Required fields are marked *