Chronic Pain Control: What’s Adequate—and Appropriate? 10 Questions Physicians Often Ask

New York University

Dr King is in private practice in pain medicine in New York and is clinical professor of psychiatry at New York University School of Medicine. He is board-certified in pain medicine and psychiatry and is a licensed acupuncturist.

ABSTRACT: The results of diagnostic tests do not correlate well with the presence and severity of pain. To avoid missing a serious underlying condition, look for “red flags,” such as unexplained weight loss or acute bladder or bowel function changes in a patient with low back pain. Nonopioid medications can be more effective than opioids for certain types of pain (for example, antidepressants or anticonvulsants for neuropathic pain and fibromyalgia). When NSAIDs are indicated, the nonacetylated NSAIDs or celecoxib may be better choices for patients who are at risk for GI problems or who are receiving anticoagulants. The tricyclic antidepressants and other serotonin-norepinephrine reuptake inhibitors are more effective as analgesics than selective serotonin reuptake inhibitors. When opioids are indicated, start with what are generally classified as the less potent agents (codeine, oxycodone, hydrocodone, oxymorphone) and then progress to stronger ones (hydromorphone, fentanyl, methadone, morphine) if needed.

Key words: chronic pain, analgesic, analgesia

Chronic pain is a significant public health problem that affects over 100 million American adults.1 Multiple guidelines have been propagated on the management of chronic pain, but many patients still suffer needlessly.

Here I answer some of the most common questions clinicians have about the evaluation and pharmacologic management of chronic pain.

1. What is the best way to assess the severity of pain?
Although pain is a subjective complaint, there are a number of different instruments for assessing its intensity.2 Among the most commonly used are:
The Numeric Pain Intensity Scale (patients are asked to rate their pain from 0 to 10, where 0 is no pain and 10 the worst possible pain).
The Verbal Pain Intensity Scale (patients are asked whether their pain is mild, moderate, or severe).
The Visual Analog Scale (patients are asked to make a mark on a 10-cm line to indicate the intensity of their pain).

These scales have been shown to be equally accurate. Select the one that you find easiest to use.

All 3 scales can be used for both acute and chronic pain; however, certain instruments have been designed specifically to evaluate chronic pain. These instruments—which include the McGill Pain Questionnaire, the Oswestry Low Back Pain Scale, and the Neuropathic Pain Questionnaire—allow for a more in-depth analysis of a patient’s pain. However, they are used more often in research than in clinical settings.

Clinicians sometimes worry that a patient may be exaggerating or even faking pain for financial gain. However, unless there is evidence that the patient is lying (as in malingering or factitious disorder), it is best to accept his or her pain complaint as real. Keep in mind that complaints of pain are common in many psychiatric disorders—most notably depressive and anxiety disorders—and that the pain associated with these disorders is as real as that secondary to physical pathology.

2. If I do not order diagnostic tests, how can I be sure that I will not miss a potentially serious condition?
Patients with chronic pain often undergo a variety of diagnostic tests in an attempt to identify the cause of their pain and to quantify their suffering. However, multiple studies have shown that there is often little correlation between the results of diagnostic tests and the presence and severity of pain. One study, for example, demonstrated that if MRI scans of the lumbar spine are performed in a randomly selected group of persons who do not have low back pain, at least 1 abnormal disk will be identified in about 66% of the participants, and problems in 2 or more disks will be evident in almost 40% of them.3 The possibility of missing a serious underlying illness is a valid concern. However, studies have shown that for the 2 most common types of pain—low back pain and headache—there are a number of “red flags” that suggest the presence of organic pathology and signal the need for diagnostic testing.

Red flags in low back pain that suggest a tumor include:
First episode of back pain at age older than 50 years in the absence of trauma.
Unexplained weight loss (most patients with chronic pain report weight gain because of inactivity).
Pain that is worse when lying down, perhaps necessitating that the patient sleep sitting up (most patients with mechanical low back pain are most comfortable lying down); this red flag is especially common in women with metastatic breast cancer.

The only condition associated with low back pain that must be treated as an emergency is cauda equina syndrome. Fortunately, this condition is relatively rare and the symptoms are usually easy to detect. These include:
Acute changes in bladder and bowel function.
Severe or progressive neurologic deficit in the lower extremities.
Perianal or perineal sensory loss.

Red flags for potentially serious conditions associated with headache include:
New onset of headache or a change in quality from previous headaches.
A headache of extraordinary severity.
Onset of headache with exertion.
Decreased alertness or cognition.
Nuchal rigidity.

Although CT and MRI scans are often included in the evaluation of patients with migraine and tension-type headaches, such studies are usually not indicated for these conditions and provide no useful information about treatment.

3. Are opioids usually the best option for patients with chronic pain?
Although the opioid analgesics are often thought of as “the best” pain medications, drugs from the other classes can provide equal or, at times, greater analgesia, depending on the cause of the pain. For example, pain that stems from bone pathology often responds better to NSAIDs than to opioids. Neuropathic pain that results from lesions in the peripheral or central nervous system is often better managed with antidepressants and anticonvulsants than with opioids. The only drugs that are FDA approved for the treatment of fibromyalgia are pregabalin, which is also approved as an anticonvulsant, and two serotonin-norepinephrine reuptake inhibitors, duloxetine and milnacipran. The triptans and ergots are used to treat migraine headaches. Finally, bear in mind that long-term opioid use can actually lower pain thresholds and exacerbate pain.4

Table 1 lists dosages and other prescribing information for the non-opioid agents most commonly used to treat chronic pain; these include NSAIDs, acetaminophen, antidepressants, and anticonvulsants. Table 2 lists prescribing information for the opioids.5,6

4. What is the optimal strategy for prescribing opioid analgesics?
A good approach is to start with what are considered less potent opioids and then progress to the stronger ones as—and if—needed.

Less potent opioids. These include the following:

Tramadol is a combination drug that contains a serotonin-norepinephrine reuptake inhibitor and a mild opioid (1/5000 the strength of morphine). Most of its analgesic effect appears to be due to the former. Tapentadol is also a combination drug that contains a norepinephrine reuptake inhibitor with an opioid receptor that has 1/50 the binding strength of morphine.

The next step up is to codeine, oxycodone, hydrocodone, and oxymorphone—all of which have comparable analgesic effects. However, individual patients may respond better to one than to the others.

Codeine, oxycodone, and hydrocodone are metabolized to their analgesic forms by the P-450 2D6 hepatic isoenzyme system. Codeine is metabolized to morphine, oxycodone to oxymorphone, and hydrocodone to hydromorphone. Thus, drugs that inhibit this isoenzyme system can markedly diminish the analgesic effects of these medications. The selective serotonin reuptake inhibitors (SSRIs), especially fluoxetine and paroxetine, are among the more widely used agents that inhibit P-450 2D6 isoenzyme.

Codeine, oxycodone, and hydrocodone are most commonly available in preparations that contain acetaminophen. Although patients can become tolerant of the analgesic effects of opioids and may in time require higher doses to manage their chronic pain, they do not become tolerant of the hepatic toxicity associated with acetaminophen. The FDA’s decision to limit the amount of acetaminophen in these combination medications to 325 mg per pill should provide for some reduction of this problem but does not eliminate it. You can avoid problems with acetaminophen toxicity by giving the opioid alone, with separate dosing of acetaminophen. Both codeine and oxycodone are available in preparations that do not include any additional medications (hydrocodone is only available in combination with other medications, such as acetaminophen or ibuprofen). Because oxymorphone is not currently available in generic form and it does not appear to be any more effective than oxycodone, which is metabolized to it, its primary advantage is for patients who are taking a P-450 2D6 inhibitor.

Oxycodone and oxymorphone are available in sustained-release preparations that can be useful for patients who require opioids for an extended period. Although there has been much publicity about the abuse of the sustained-release form of oxycodone, it does not appear to be associated with more problems than the other opioids, if it is used correctly.

More potent opioids. These include the following:

Meperidine is generally contraindicated in all forms of chronic pain. Its metabolite, normeperidine, has a longer half-life than meperidine and therefore can accumulate with repeated dosing. Normeperidine is a cerebral irritant that can cause seizures and death. Meperidine also has the disadvantage of poor oral bioavailability.

Hydromorphone, morphine, fentanyl, and methadone can all be beneficial. Hydromorphone and morphine are available in both short- and long-acting preparations. Fentanyl is available in preparations that are absorbed transdermally or transmucosally. These can be especially useful for patients who have difficulty in tolerating oral medications, such as cancer patients with nausea and vomiting. The transdermal preparation of fentanyl is usually applied for 72 hours and is indicated for chronic pain; the transmucosal form is used to manage breakthrough pain and not as the primary opioid analgesic.

Although methadone is usually regarded as an agent used to treat opioid addiction, it is an important analgesic. In addition to acting on the opioid receptors in the CNS, it also appears to provide analgesia by blocking the N-methyl-D-aspartate (NMDA) receptors. Although methadone has a long half-life (which is why it can be given once a day for the treatment of opioid addiction), its analgesic effect is more short-lived. Thus, for optimal analgesia, methadone must be given every 6 to 8 hours. When such a dosing schedule is used, methadone provides a stable level of analgesia comparable to that of the other longer-acting opioids—including controlled-release morphine.

There is no ceiling dosage for the µ-opioid receptor agonists (which include morphine, methadone, fentanyl, hydromorphone, codeine, oxycodone, oxymorphone, and hydrocodone). Because patients may develop tolerance to the analgesic effects of these medications, increasingly higher doses may be needed over time to achieve the same level of pain relief.

If a patient requires more than 2 doses per day of an immediate-release opioid for an extended period, consider prescribing a long-acting opioid. These medications provide more stable analgesia over the course of a day than short-acting opioids. However, long-acting opioids must be taken on a fixed schedule to be effective. Long-acting opioids are not useful for acute pain because there can be a delay of several days before their maximum analgesia is achieved.

When you switch from a short-acting to a long-acting opioid, dose the long-acting medication conservatively at the beginning. This is especially important when a long-acting formulation of a different opioid is used.

In addition to the long-acting opioid, prescribe a short-acting one initially on an “as-needed” basis for breakthrough pain. Select the short-acting opioid based on the potency of the long-acting opioid. For example, hydromorphone can be useful for breakthrough pain in patients who take long-acting hydromorphone, methadone, morphine, or fentanyl. Short-acting oxycodone and oxymorphone are usually prescribed for breakthrough pain in patients who take the long-acting formulation of the same drugs. Knowing the amount and frequency of breakthrough medication a patient uses can help you adjust the dosage of the long-acting opioid. After the first 1 to 2 weeks of use of a long-acting opioid, this drug should provide most of the analgesia the patient needs, with breakthrough medication required relatively infrequently and usually only during activities that acutely exacerbate the pain.

5. Is it legally safe to prescribe opioids for chronic pain?

The use of opioids for chronic pain is legally safe as long as the medications are prescribed in accordance with the law and with accepted medical practice. Many states have adopted all or part of the model guidelines on the use of controlled substances for pain issued by the Federation of State Medical Boards of the United States.7 These guidelines clearly note that opioid analgesics have a legitimate place in the management of pain and that extended use may be required. The number of physicians who are disciplined by the DEA for prescribing opioids is exceedingly small, and there is no evidence that physicians who are prescribing them in good faith and following the law have anything to fear from legal authorities.8

A growing public health issue in this country is the use of prescription opioids for non-medical reasons. Unfortunately, until relatively recently many medical professionals maintained that patients who were started on opioids for legitimate purposes never ended up abusing or becoming addicted to these medications, although as long as 20 years ago there was already evidence that a significant number of these patients developed these problems. Bouckoms and colleagues9 studied 59 patients with chronic noncancer pain treated with opioid analgesics for an average of 36 months. They found that 24% of the participants became addicted to these medications. However, the authors were unable to discern any factors that could help predict which patients were more likely to have problems with opioids.

In my practice, it is not uncommon to see patients with chronic pain who either abuse or become addicted to opioid analgesics. This is not a reason to avoid prescribing opioids for pain. However, it does indicate that addiction is a potential problem. Monitor all patients who are being treated with opioids for signs of abuse or addiction. Such signs include10:
Obtaining opioids from several physicians.
Repeatedly using doses higher than those prescribed. A clue to this is that the patient runs out of the medication before the next scheduled prescription.
Hoarding the medication during periods when the pain is less severe.
Abusing other substances, such as alcohol or illicit drugs.

6. Which antidepressants have the greatest analgesic effect?
Antidepressants have been shown to provide analgesia for a wide range of chronically painful conditions, including neuropathic pain (such as that associated with diabetic neuropathy), osteoarthritis, rheumatoid arthritis, and irritable bowel syndrome. Antidepressants are particularly efficacious in the management of fibromyalgia and can provide prophylaxis for migraine and tension-type headaches as well.11

Antidepressants that affect both the serotonergic and noradrenergic systems appear to provide the most analgesia. Thus, the tricyclic antidepressants (TCAs), venlafaxine, and duloxetine, which block the reuptake of both serotonin and norepinephrine, are generally much more effective as analgesics than SSRIs such as fluoxetine, paroxetine, sertraline, and citalopram. Moreover, when TCAs are administered to patients who are also taking opioids, they appear to bolster the analgesic effect of the opioids.

How antidepressants effect analgesia is unclear. Among the suggested mechanisms of action are:
Enhanced endorphin secretion.
Sodium channel blockade.
Inhibition of substance P, a neuropeptide involved in both pain and depression.
NMDA receptor antagonism. 

The analgesic effect of antidepressants seems to be separate from their antidepressant effect; thus, patients with chronic pain do not need to be depressed in order to benefit from the analgesia these agents provide; in fact, they have no psychiatric effect on patients who are not depressed.

All the TCAs appear to be equally analgesic. Although amitriptyline is commonly thought to be either the only TCA that provides analgesia or the TCA that provides the most analgesia, neither claim is substantiated. Amitriptyline may have been considered most beneficial because of its strong antihistaminic effect, which makes it the most sedating of the TCAs. Many patients with chronic pain have difficulty in sleeping and so are helped by this sedating effect. However, patients may find amitriptyline too sedating. In addition, this agent has the strongest anticholinergic effects of the TCAs; it can cause constipation and urinary retention. Other TCAs, such as desipramine and nortriptyline, have more benign side-effect profiles.

Because venlafaxine and duloxetine do not have the anticholinergic effects of the TCAs or affect cardiac functioning as TCAs can, these agents can be useful alternatives. To achieve sufficient analgesia, 100 to 150 mg/d of amitriptyline, an equivalent dosage of another TCA, or 150 to 300 mg/d of extended-release venlafaxine is required in most patients.

The only FDA-approved antidepressant that is also approved as an analgesic is duloxetine, which is approved for the treatment of fibromyalgia, diabetic peripheral neuropathic pain, and chronic pain associated with chronic osteoarthritis and chronic low back pain. It is usually recommended that dosing of duloxetine begin at 30 mg/d for 1 week to diminish the possibility of nausea, a common side effect. The dose is then increased to 60 mg q/d. There is limited evidence that increasing dosages beyond this will have much more of an analgesic effect, but there do appear to be patients who may benefit from a higher dose of 90 or even 120 mg/d. Although milnacipran, which is FDA approved for fibromyalgia, is marketed overseas as an antidepressant, it is not approved for treatment of depression in the United States.

7. Which anticonvulsants provideanalgesia?
Most anticonvulsants appear to provide analgesia and are especially effective for neuropathic pain.12 The anticonvulsants approved by the FDA for the management of pain are:
Gabapentin, which is indicated for postherpetic neuralgia.
Pregabalin, which is approved for postherpetic neuralgia, diabetic neuropathic pain, and fibromyalgia.
Topiramate, which is approved for the prophylaxis of migraine headaches.
Carbamazepine, which is approved for the treatment of trigeminal neuralgia.

Because the new anticonvulsants have more benign side-effect profiles, they have generally replaced the older anticonvulsants (such as phenytoin, carbamazepine, and valproate sodium) for use in pain management. There is one notable exception: for reasons that are still unknown, trigeminal neuralgia usually responds better to carbamazepine than to any other medication.

How anticonvulsants exert their analgesic effect is unclear. It is thought that neuropathic pain may be to some degree a result of destabilization of the neuronal membranes and that anticonvulsants restabilize the membranes by blocking the calcium and sodium channels. Some of the analgesic properties of gabapentin may be related to increases in the synthesis and release of gamma-aminobutyric acid in the brain caused by this agent.

8. How does one choose an NSAID?
All the NSAIDs appear to be equally analgesic; thus, the selection should primarily be based on their side-effect profiles. Although a great deal of research still remains to be performed on this subject, there is currently sufficient information to provide some general recommendations.

For years the primary adverse effects of concern that were associated with the use of NSAIDs were GI toxicity and inhibition of antiplatelet activity, which increases the risk of bleeding. These, especially the apparent reduction in GI toxicity, rather than any difference in efficacy, were the major reasons why the selective cyclooxygenase-2 (COX-2) inhibitors attracted such attention when they were first introduced. Although both of these adverse effects remain significant concerns, other side effects, especially cardiovascular problems, also need to be considered when deciding which NSAID to use. As a result of these other side effects, two of the three COX-2 inhibitors that were approved by the FDA are no longer available, leaving only celecoxib on the market in the United States.

With regard to GI toxicity, celecoxib does appear to be less likely than the other currently available NSAIDs to cause this effect, but it is still a significant problem with this drug. As to inhibition of platelet function, the nonacetylated salicylates such as salsalate and choline magnesium trisalicylate appear to be less likely to be associated with this effect.

As research has focused on cardiovascular problems associated with NSAID use, we have become aware that it is not only the COX-2 inhibitors but, to varying degrees, all NSAIDs that are associated with these problems. Despite the focus on the GI effects of NSAIDs, it has become apparent that the cardiovascular problems are also frequent and should be weighed in the decision as to which NSAID to use. Of the most commonly prescribed NSAIDs, naproxen appears to be the safest with regard to cardiovascular adverse events, including risk of stroke and myocardial infarction.13

Several NSAIDs are available in topical preparations. These can provide substantial analgesia with a reduced risk of systemic side effects compared to the oral NSAIDs.14 However, whether they are as efficacious as oral NSAIDs is yet to be

Because it is impossible to predict which NSAID a patient will respond to best, if a trial of one is not effective, it does not mean that the patient will not find another more efficacious. It is also important to remember that acetaminophen is often a practical alternative to NSAIDs; this drug is usually a safer choice than any of the NSAIDs in patients without hepatic disease as long as they are careful not to exceed the daily dosage limit.

9. Are there any other medications that arehelpful in the management of chronic pain?
The lidocaine patch is useful in the treatment of localized pain. It is only FDA-approved for the management of postherpetic neuralgia; however, studies indicate that it can also provide significant relief for other types of pain, including low back pain.15 I have used lidocaine patches in a variety of neuropathic pain syndromes, including diabetic neuropathy and complex regional pain syndrome (reflex sympathetic dystrophy); I have also found them to be effective in musculoskeletal pain in some patients. Lidocaine patches are left on for 12 hours per day; up to 3 patches can be applied at a time. This medication has a very benign side-effect profile because the analgesic effect is local rather than central. Thus, it is an especially good choice for patients who may have difficulty in tolerating other analgesics.

Capsaicin, a topical medication derived from chili peppers, is useful in the treatment of osteoarthritic pain, localized musculoskeletal pain, and neuropathic pain. Although it usually feels warm when applied, capsaicin’s analgesic effect is not a result of the counterstimulation associated with some other topical agents. Its effect has primarily been viewed as secondary to the depletion of peripheral substance P, a neuropeptide involved in the transmission of pain; however, it may actually be the result of a neurotoxic effect on nociceptors.

Capsaicin is available over the counter in topical cream preparations in 0.025% and 0.075% strengths and in a patch with a concentration of 0.9%. I generally recommend the cream rather than the patch because the former is less expensive. Side effects do not appear to be dose-related; however, the higher concentrations may produce a stronger burning sensation. Thus, I recommend starting with the 0.075% preparation. Capsaicin is usually applied 3 or 4 times a day. In my experience, the lidocaine patch tends to be more effective than capsaicin.

One class of medications that are often used inappropriately in chronic pain are the benzodiazepines. Benzodiazepines are generally contraindicated in the treatment of chronic pain—for several reasons:
Extended use of benzodiazepines can actually lower pain thresholds and thereby increase pain.
Benzodiazepines can interfere with the analgesic effects of opioids.
Benzodiazepines are potentially physically and psychologically habituating; thus, it can be difficult to discontinue their use.

The 2 settings in which benzodiazepines are most often used in patients with chronic pain are insomnia and spasm. Other medications are superior choices for these problems. Insomnia can be better treated with either a sedating antidepressant or with medications specifically formulated for insomnia (for example, zolpidem or zaleplon); spasm is better treated with cyclobenzaprine.

What nonpharmacologic modalities are useful for pain management?
Among the most commonly used are physical therapy and occupational therapy. Patients with chronic pain are often in poor physical condition because the pain impedes their ability to exercise, and the resultant inactivity can, in turn, exacerbate the pain. Physical therapy can frequently improve conditioning and teach patients better body mechanics. Occupational therapy can improve patients’ performance of activities of daily living.

Psychotherapeutic techniques can also be beneficial. Many patients with chronic pain have comorbid depression, which may respond to psychotherapy. Therapies that focus on the pain—most notably relaxation techniques and biofeedback—can also be helpful.

Acupuncture can also provide analgesia in a variety of pain conditions.16 Although the pain relief produced by acupuncture is often dismissed as a placebo effect, acupuncture appears to increase the release of endorphins and the bioavailability of serotonin and norepinephrine. Acupuncture has virtually no adverse effects when it is performed by a trained professional and disposable needles are used. ■


1. Institute of Medicine. Relieving Pain in America. Washington, DC: Institute of Medicine; 2011.
2. Phillips KA, Fallon BA, King SA. Somatoform and factitious disorders and malingering measures. In: Rush AJ, First MB, Blacker D, eds. Handbook of Psychiatric Measures. 2nd ed. Washington, DC: American Psychiatric Publishing; 2008:559-585.
3. Jensen MC, Brant-Zawadzki MN, Obuchowski N, et al. Magnetic resonance imaging of the lumbar spine in people without back pain. N Engl J Med. 1994;331:69-73.
4. Inturrisi CE. Clinical pharmacology of opioids for pain. Clin J Pain. 2002;18:S3-S13.
5. American Geriatrics Society Panel on the Pharmacological Management of Persistent Pain in Older Persons. Pharmacological management of persistent pain in older persons. JAGS. 2009;57:1331-1346.
6. Chou R, Fanciullo GJ, Fine PG, et al. American Pain Society-American Academy of Pain Medicine Opioids Guidelines Panel. Clinical guidelines for the use of chronic opioid therapy in chronic non cancer pain. J Pain. 2009:10:113-130.
7. Federation of State Medical Boards of the United States. Model  Policy for the Use of Controlled Substances for the Treatment of Pain. Dallas: Federation of State Medical Boards; 2004.
8. Jung B, Reidenberg MM. The risk of action by the Drug Enforcement Administration against physicians prescribing opioids for pain. Pain Med. 2006;7:353-357.
9. Bouckoms AJ, Masand P, Murray GB, et al. Chronic nonmalignant pain treated with long-term oral narcotic analgesics. Ann Clin Psychiatry. 1992;4:185-192.
10. Chou R, Fanciullo GJ, Fine PG, et al. Opioids for chronic noncancer pain: prediction and identification of aberrant drug-related behaviors: a review of the evidence for an American Pain Society and American Academy of Pain Medicine clinical practice guideline. J Pain. 2009;10:147-159.
11. Hauser W, Bernardy K, Uceyler N. Treatment of fibromyalgia syndrome with antidepressants.JAMA. 2009;301:198-208.
12. Finnerup NB, Sindrup SH, Jensen TS. The evidence for pharmacological treatment of neuropathic pain. Pain. 2010;150:573-581.
13. Trelle S, Reichenbach S, Wandel S, et al. Cardiovascular safety of non-steroidal anti-inflammatory drugs: network meta-analysis. BMJ. 2011;342:C7086.
14. Makris UE, Kohler MJ, Fraenkel L. Adverse effects of topical nonsteroial antiinflammatory drugs in older adults with osteoarthritis: a systematicliterature review. J Rheumatol. 2010;37:1236-1243.
15. Dworkin RH, Jensen MP, Gould E, et al. Treatment satisfaction in osteoarthritis and chronic low back pain: the role of pain, physical and emotional functioning, sleep, and adverse events. J Pain. 2011;12:416-424.
16. Chou R, Huffman LH. Nonpharmacologic therapies for acute and chronic low back pain: a review of the evidence for an American Pain Society/American College of Physicians clinical practice guideline. Ann Intern Med. 2007;147:492-504.