Letter to the Editor

Two readers express concerns about pain management recommendations that appeared in our April cover article, and a story source responds.


We can identify with the challenges surrounding the management of pain. However, we are concerned about some of the recommendations in the recent article “The prescription for pain management” (ACP Hospitalist, April 2013).

For conversion from an intravenous to an oral opioid regimen, the article recommends that 40% of the calculated 24-hour oral morphine equivalent be given in an immediate-release form on an as-needed (PRN) basis and the other 60% be scheduled as long-acting. This could risk gross underdosing, which may contribute to uncontrolled pain. This, in turn, could lead to behaviors that may be falsely interpreted as drug-seeking when, in fact, they are a result of ineffective pain management (pseudoaddiction). Furthermore, the article's suggestion offers the novice no guidance or instruction for adjusting PRN dosing if the initial dose is ineffective. Our standard of practice is to use the total 24-hour oral morphine equivalent as a guide to the PRN dosage and frequency. Restricting the total PRN use to the remaining 40% limits the patient. Standard of care is to prescribe 60% (or, in our practice, 50%) of total daily use as the basal level of medication and allow the patient to use as much immediate release as needed to achieve pain control. The patient may have more pain on post-op day two than on day one, and limiting the PRN dose to 40% of day one use would be ineffective for management.

The article also recommends long-acting opioids after procedures for patients experiencing acute pain syndromes. This risks overestimation of pain, which could easily result in poor patient outcomes. In our experience, the standard of practice for treatment of acute pain in opioid-naive patients is with immediate-release formulations only. These individuals' pain is expected to resolve over a relatively short period, and a long-acting formulation creates concerns about sedation, respiratory depression and death. Moreover, the practical benefit of long-acting medications is for dosing convenience for predictable pain syndromes (i.e., the patient takes the drug less often). When pain syndromes are unpredictable, as postoperatively when pain intensity and frequency are likely to decrease daily, scheduling long-acting medications is not recommended. Rather, we allow for appropriate PRN dosing, the right dose and frequency, and allow the patient to use as-needed medications to achieve pain control. If, at follow-up, opioid use is still significant, reevaluation is warranted for chronic pain or procedure-related complications.

The unanticipated consequences of the article's recommendations are twofold. First, scheduling a long-acting opioid in acute and resolving pain syndromes may mask postoperative complications. Second, the patient may develop dependence/addiction if not followed closely and if the medication is not tapered and discontinued promptly as the acute pain resolves.

Beth Clark, CNP
Kavitha Norton, DO, ACP Member
Columbus, Ohio

In response: Ms. Clark and Dr. Norton raise reasonable questions that bear consideration and require further discussion. As I address them, I will also clarify some clinical assertions that may not be clearly detailed in the original article. I believe that many of the voiced concerns are based on misunderstanding.

Ms. Clark and Dr. Norton take issue with the proposed dose limits for PRN medication when converting from intravenous to oral regimens, suggesting that the dosage should be open-ended and unrestricted. I assert that the goal of establishing a 24-hour opioid requirement is to make a safe and educated guess of what the patient will likely need while minimizing the risk of unintended side effects or harm due to overdose. Open-ended PRN dosing creates a situation where the patient has access to potentially far too much medication, all in the context of less surveillance (i.e., the patient would have access to untested amounts of medication while at home or alone, instead of while in the hospital being monitored by a nurse).

Furthermore, it is certainly true that a patient's pain levels and medication requirements may change from day to day. The article doesn't suggest that the first postoperative day's pain medication be used as a guide. Rather, it means to recommend assessing the patient's pain medication needs on a day closer to discharge (such as the second or third postoperative day), which is likely to better approximate the patient's ongoing needs as an outpatient during healing. As with all pain situations, the best practice is to observe the patient's needs and response to a prescribed regimen and then use these clinical assessments to make informed decisions to increase or decrease the dosing (PRN or otherwise). Outside a monitored health care setting, access to nonmedically supervised, limitless PRN dosing is not safe.

Ms. Clark and Dr. Norton also believe that the article recommends “long-acting opioids after procedures for patients experiencing acute pain syndromes.” This is not the case. I do agree, in principle, that acute pain syndromes (thought to have a short and time-limited recovery period) should not typically require long-acting opioid regimens. The use of long-acting opioids should be based on the extensiveness of the injury, the anticipated recovery period, and the dose required to manage the patient's pain. Consequently, long-acting opioids are typically reserved for situations requiring significant doses for extended periods. This point could have been fully expounded on in the article, however, since the lone “sample calculation” included does indeed show a conversion to an oral long-acting and short-acting opioid formulation (it is mathematically accurate, and there was no suggestion of using this calculation for all postoperative situations). Very plainly, the article does not suggest routinely starting long-acting opioids for all postprocedure pain or for acute pain syndromes. The choice of these agents always depends on the clinical scenario.

I thank Ms. Clark and Dr. Norton for their thoughtful response. The article's aim was not to speak comprehensively and authoritatively about all pain scenarios, as this is the domain of textbooks, fellowships and prolonged clinical experience. Rather, it intended to give the reader some perspective and insight into common pain strategies that may be helpful and safe for patients. There are definite subtleties to practicing pain medicine, and there exist exceptions to every rule. Pain management is complicated and always requires careful deliberation of risks and benefits, along with thorough understand of each patient's specific clinical situation.

Jason S. Eldrige, MD
Rochester, Minn.