In the last few decades, the use of daily opioids to treat chronic nonmalignant pain has increased (1,2). Whether long-term daily opioid use results in sustainable pain relief for chronic nonmalignant pain conditions is debatable (3–5). Some patients require rapid dose escalation to maintain opioid efficacy, suggesting that opioid tolerance development may be a clinical problem. For example, multiple studies of chronic administration of oral or intrathecal opioids suggest that, whereas prolonged pain relief can be obtained in many patients, there are patients with nonmalignant pain who tend to require increasing doses of opioids over time (6–8). Paice et al. (7) demonstrated a 600% increase in intrathecal morphine dosing in nonmalignant pain patients over a 2-yr period (although catheter fibrosis may have confounded these data). Mystakidou et al. (8) report patients escalating from 75 μg/h or less of transdermal fentanyl patches up to 250 μg/h within 18 mo.
Opioid escalation can occur for a variety of reasons, including underlying disease progression, opioid addiction, and tolerance development. We have diagnostic tools to help identify disease progression, and there are guidelines in the literature to identify and manage pain patients who might be drug seeking or have a history of substance abuse (9,10). However, there are no guidelines to identify patients who may be poor candidates for long-term opioid treatment because of rapid opioid tolerance development and unsustainable long-term pain relief.
Whereas there are numerous reports studying the effects of age on pharmacokinetic variables of opioids, clinical pharmacodynamic tolerance to long-term daily opioids has never been studied in an age-dependent manner. Most clinical studies analyze patients from 18–80 yr as a single group, with the mean age usually in the 50- to 60-yr range. A study published in 1975 (11) suggested that the development of tolerance to daily morphine administration occurs more rapidly in young rats; however, the oldest rats used in this study were 12 wk. We completed a study of daily morphine administration in rats ranging in age from 3 wk to 1 yr (12). We found a dramatic prolongation in the time to onset of tolerance as the rats aged. These findings prompted us to perform a retrospective study of our pain clinic patients to determine whether opioid dose escalation was significantly different based on the age of the patient, as well as other factors, such as type of pain. Also we were interested in determining whether treatment with chronic long-acting opioids resulted in a significant reduction in pain scores over time.
With approval from our IRB, we had access to approximately 10,000 clinic charts for patients who entered our clinic between 1988 and 2001. Charts were randomly chosen from both the clinic and chart storage facility based on chart identification number until our goal of approximately 100 patients in each age group was achieved. Patients 50–60 yr were excluded from the study to more definitively separate the average age in the younger and older groups. To be included in the study, patients had to have nonmalignant pain that was treated either without opioids or only small-dose, short-acting opioids before the time of their first appointment in our clinic. Long-acting opioids had to have been started and maintained under pain clinic supervision for at least 6 mo. Approximately 950 charts were analyzed by using the previously described method until we obtained 104 patients younger than 50 yr and 102 patients aged 60 yr and older that met the entrance criteria. This group of 206 patients was treated by 11 different faculty physicians, thereby reflecting a wide variety of pain management practice styles. In addition to initiation of daily long-acting opioids, a multi-disciplinary approach was used for all patients, including psychological assessment and treatment if required, physical therapy as indicated, use of adjuvant medications (including antiinflammatory drugs, antidepressants, anticonvulsants, muscle relaxants, etc.), and interventional procedures when indicated. No patients implanted with a spinal cord stimulator or intrathecal pump were included in this study. For study analysis, all opioid dosing was converted to oral morphine sulfate equivalent (milligrams) per day based on published equianalgesic conversion tables (13–16). Daily opioid dose was recorded at the initiation of long-acting opioid therapy, and the peak opioid dose and the final opioid dose were recorded at patient discharge from the clinic. Visual analog scale (VAS) pain scores were recorded upon initial clinic visit, just before starting daily long-acting opioid therapy and at the final visit before discharge.
Oral morphine equivalent doses were compared using the nonparametric Mann-Whitney U-test and are reported as mean ± se of the mean (sem), and also the median is reported. The VAS scores were analyzed using the repeated-measures analysis of variance with Bonferroni post hoc test.
The patient demographics, number of months of chronic pain before clinic admission, type of pain, and percentage of patients in each group on short-acting opioids at time of admission to the pain clinic are shown in Table 1. The nociceptive pain category consisted of patients with no identifiable neuropathic component (e.g., fibromyalgia or arthritis), whereas the neuropathic category consisted of patients with an identifiable neuropathic component with or without a nociceptive component (e.g., postherpetic neuralgia or low-back pain with radiculopathy). In the older-age group, most (>60%) nociceptive pain consisted of arthritis (facet, knee, shoulder, etc.). Much less common was myofascial pain (15%) and headache (10%), with all other diagnoses infrequent. Neuropathic pain conditions in the older group consisted of approximately 40% radiculopathy, 20% cranial or peripheral nerve injury (such as trigeminal neuralgia and phantom limb pain), 15% postherpetic neuralgia, and 15% peripheral neuropathy. In the younger group, arthritis accounted for approximately 40% of the nociceptive pain conditions, followed by myofascial pain (20%), pelvic or abdominal pain (20%), and headache (15%). Neuropathic pain conditions were mainly radiculopathies (50%), followed by cranial or peripheral nerve injury (20%), peripheral neuropathies (15%), and complex regional pain syndrome (10%). Other diagnoses were uncommon in occurrence.
Daily oral morphine-equivalent dose at initiation, peak opioid dose, and opioid dose at time of discharge from the clinic are graphed for both the younger and older groups in Figure 1. There was no significant difference in initial opioid dosing between younger (49 ± 3 mg/d; median, 37 mg/d) and older (42 ± 3 mg/d; median, 30 mg/d) age groups. Both age groups increased opioid dosing over the same time period (peak opioid dose reached at 15.0 ± 1.3 mo in the younger group versus 14.4 ± 1.5 mo in the older group). However, the average peak opioid dose achieved in the younger group (452 ± 63 mg/d; median, 301 mg/d) was more than two-fold the maximum average dose achieved in the older group (211 ± 23 mg/d; median, 150 mg/d; P < 0.0001). Therefore, the younger group escalated on average 27 mg of daily oral morphine equivalent per month during the 15-mo period, whereas the older group escalated at a substantially slower rate of 12 mg of daily oral morphine equivalent per month. The total time on long-acting opioids (at last clinic appointment) for both age groups was similar and averaged 26.5 ± 1.6 mo in the younger group and 27.5 ± 2.1 mo in the older group. The final opioid dose was 20% smaller than the peak dose in both age groups, reflecting a halt and slight reversal of the dose escalation seen in the initial time period, which probably reflects physician-imposed prescribing limits or opioid-induced side effects. Final opioid dosing at time of discharge from the clinic remained significantly different among the age groups (younger, 365 ± 61 mg/d; median, 221 mg/d; older, 168 ± 18 mg/d; median, 101 mg/d; P < 0.0001).
There were no differences in opioid dosing based on sex in either age group at initiation of long-acting opioid therapy, peak opioid dosing, or final opioid dosing (data not shown).
Within each group, a similar analysis was performed taking into account the patient’s type of pain: nociceptive versus neuropathic (Fig. 2, A and B). In the older patient population, no difference in opioid dosing was apparent based on type of pain (Fig. 2A). The older nociceptive pain group initiated long-acting opioid therapy at an average dose of 35 ± 4 mg/d (median, 30 mg/d) of oral morphine equivalent and increased to an average peak of 238 ± 40 mg/d (median, 175 mg/d) over 18.1 ± 3.3 mo. The older neuropathic pain group initiated long-acting opioid therapy at an average dose of 46 ± 4 mg/d (median, 37 mg/d) of oral morphine equivalent and increased to an average peak of 198 ± 28 mg/d (median, 150 mg/d) over 12.6 ± 1.6 mo. Therefore, the older nociceptive group averaged 11 mg daily of oral morphine equivalent per month escalation, whereas the older neuropathic pain group averaged 12 mg daily of oral morphine equivalent per month escalation. The final dose in the older nociceptive group averaged 200 ± 38 mg/d (median, 90 mg/d) after 29.4 ± 3.7 mo on long-acting opioids and was not significantly different from the final dose in the older neuropathic group, which averaged 153 ± 18 mg/d (median, 102 mg/d) after 26.6 ± 2.6 mo on long-acting opioids.
However, in the younger patients, whereas the opioid dose was similar at initiation of therapy (nociceptive, 55.1 ± 5.3 mg/d [median, 37 mg/d]; neuropathic, 42.4 ± 4.3 mg/d [median, 35 mg/d]), the dose escalation was significantly less (16 mg daily of oral morphine equivalent per month) in the neuropathic group (maximum dose, 317.9 ± 36.8 mg/d [median, 244 mg/day] over 17.2 ± 2.3 mo) than the nociceptive group (38 mg daily of oral morphine equivalent per month) (maximum dose, 562.9 ± 108.9 mg/d [median, 369 mg/d] over 13.2 ± 1.3 mo; P = 0.013) (Fig. 2B). The maximum dose in the younger neuropathic group (317.9 mg/d) was still significantly larger than the older group’s maximum dose of 210.8 mg/d (P = 0.002). The final dose in the younger nociceptive group averaged 449.4 ± 106.3 mg/d (median, 270 mg/d) after 25.2 ± 2.1 mo on long-acting opioids, whereas the final dose in the younger neuropathic group averaged 263.5 ± 36.9 mg/d (median, 175 mg/d) after 28.2 ± 2.4 mo on long-acting opioids (P = 0.06).
VAS scores were averaged for each age group upon first clinic appointment, before the first dose of long-acting opioid therapy, and at the time of discharge from the clinic on long-acting opioids (Fig. 3, A and B). Both age groups had a significant reduction in their VAS scores from first clinic appointment up until starting long-acting opioid therapy (younger, 8.4 ± 0.2 versus 6.2 ± 0.3; P < 0.0001; older, 8.3 ± 0.2 versus 6.9 ± 0.3; P < 0.01). The younger group averaged 3.9 ± 0.6 mo from the first clinic appointment until starting long-acting opioids, whereas the older group averaged 6.87 ± 1.1 mo.
Important to note is that the younger group did not have a change in VAS scores measured from start of long-acting opioid therapy until time of discharge from the clinic (6.2 ± 0.3 versus 6.1 ± 0.3), although there was a 640% (49.3 mg/d to 365.4 mg/d) increase in daily opioid dose from the initiation of long-acting opioids until discharge on opioids—a period of 26.4 mo. In the older group, once long-acting opioid therapy was initiated, VAS scores further decreased from 6.9 ± 0.3 to 5.6 ± 0.3 (P < 0.01).
There were no differences in VAS scores depending on type of pain within each age group at any time point (from the first appointment, before starting the long-acting opioid, to discharge from the clinic. There were no differences in VAS scores based on sex in either age group.
Our study found that older patients, regardless of sex or type of pain, increased opioid use significantly less than younger patients over the two-year treatment period. Most notably, younger patients with nonneuropathic pain escalated at a tremendous rate for more than a year. Based on the rate of escalation and the lack of benefit reflected by VAS scores, the large doses achieved by younger patients were limited, based on either the physician’s willingness to continue dose escalation or the patient’s reporting of intolerable side effects, and not the result of attainment of prolonged opioid efficacy.
In agreement with this finding is the study by Moulin et al. (17) who studied 46 chronic pain patients with nonneuropathic pain taking sustained-relief oral morphine up to 60 mg twice a day. This is one of the few published chronic pain studies in which the patients averaged only 40 years of age; therefore, this group is almost identical to our younger nonneuropathic group. Once the three-week dose-escalation phase was completed, patients during the 6-week evaluation on stable morphine dosing reported an almost complete loss of pain relief. Patients showed no functional improvement on opioid therapy in this study.
It is noteworthy that we did not include functional outcome measures in our study because there was not a consistent objective reporting of functional variables in the records. It could be argued that although no change was observed in VAS scores, the younger patients may have been more functional on the large-dose opioids. However, studies assessing VAS scores and functional outcome measures with opioid therapy for chronic nonmalignant pain have shown that increased functional outcomes are correlated with a decrease in VAS scores (6,18). Furthermore, the older patients in our study demonstrated long-term analgesic efficacy with opioids, based on VAS assessment, with less than half the dose escalation. Therefore, it is more likely that the younger patients experienced more rapid tolerance development resulting in no change in VAS scores rather than assuming an age-biased effect on pain reporting that is not substantiated in the literature.
It is unlikely that a difference in the rate of disease progression between the two age groups accounts for the differences in opioid escalation. Both age groups had a similar duration of chronic pain (more than eight years; Table 1) before entrance into the clinic, and patients with cancer pain were excluded from the study. Although age group entrance characteristics were similar between the two groups (Table 1), there was a larger ratio of neuropathic pain patients in the older study population compared with the younger group (Table 1). Because type of pain did not have an effect on rate of opioid escalation or VAS scores in the older population, this discrepancy does not seem to have altered the outcomes of this study.
The slower rate of opioid escalation observed in younger patients with neuropathic pain compared with younger patients with nociceptive pain may have been due to multiple factors. There are published animal data that support the concept of slower opioid tolerance development in neuropathic pain states compared with nociceptive inflammatory pain conditions (19,20), although no clinical studies have definitively substantiated this finding. It may also be the case that although opioids have proven efficacy in neuropathic pain conditions (21), more emphasis is placed on the aggressive use of adjuvant analgesics, such as anticonvulsants and tricyclic antidepressants, and therefore, opioid escalation is not aggressively pursued by either physician or patient. Yet, another factor may be the patient’s perception that their neuropathic pain is not effectively managed with opioids, and therefore, the lack of escalation is more because of a lack of perceived benefit by patient or physician. Finally, regardless of the type of pain, we cannot exclude the effect of pharmacokinetic differences between the two age groups.
If rapid tolerance to daily opioids occurs in younger chronic pain patients, is there a rationale for simply continuing to increase the opioid dose? Even with rapid tolerance development, a decrease in the younger group’s VAS scores may have been observed if we had continued opioid escalation throughout the entire two-year period of opioid therapy in our clinic. Studies reporting favorable analgesic data in long-term treatment of nonmalignant pain with opioid therapy have some patients escalating to half a gram per day of oral morphine equivalent or larger (conversion from intrathecal dose) (6–8). Long-term consequences of large-dose opioid therapy are receiving more attention lately (22), and with these concerns, the rationale for rapid chronic opioid escalation in nonmalignant pain comes into question.
The efficacy of chronic opioid therapy in the older group is an important finding of this study. The initial benefits of the multi-disciplinary pain management approach were enhanced by the addition of chronic long-acting opioid therapy. Long-term stable dosing was observed over two years accompanied by a significant reduction in VAS scores. The molecular changes that occur in neurons with aging are an important area of research that may provide novel insights into molecular mechanisms of opioid tolerance development.
The retrospective nature of this study is an obvious limitation, and a follow-up prospective study would be optimal. We can only suggest exercising caution when considering chronic daily opioids in younger patients with nonmalignant pain, especially for nonneuropathic pain conditions. Although opioid rotation is an option in cancer pain (23), it is unclear how effective or feasible opioid rotation is over decades of treatment. In some younger patients, focusing on the use of non-opiate therapies, along with intermittent use of opioids for severe breakthrough pain, has provided effective long-term pain management (anecdotal clinical observations). The use of opioids for acute exacerbation of chronic pain in younger patients can minimize their anxiety regarding these episodes as well as reduce their visits to the emergency room. Ultimately, more attention needs to be focused on the development of effective analgesics with minimal tolerance development.
In summary, age seems to be an important factor in determining which patients will require escalating doses of chronic daily opioids over time. Patients younger than 50 years escalated oral opioids at more than twice the rate of patients older than 60 years during the first year of opioid therapy. Although both age groups reported decreased pain scores after enrolling in our pain management center, only the older patients reported an additional decrease in pain scores two years after starting chronic daily long-acting opioid therapy.
1. Bell JR. Australian trends in opioid prescribing for chronic non-cancer pain, 1986–1996. Med J Aust 1997;167:26–9.
2. Clark JD. Chronic pain prevalence and analgesic prescribing in a general medical population. J Pain Symptom Manage 2002;23:131–7.
3. Foley KM. Clinical tolerance to opioids. In: Basbaum AI, Besson JM, eds. Towards a new pharmacotherapy of pain. Chichester, UK: John Wiley & Sons Ltd, 1991:181–203.
4. Savage SR. Opioid therapy of chronic pain: assessment of consequences. Acta Anaesthesiol Scand 1999;43:909–17.
5. Attal N, Guirimand F, Brasseur L, et al. Effects of IV morphine in central pain: a randomized placebo-controlled study. Neurology 2002;58:554–63.
6. Zenz M, Strumpf M, Tryba M. Long-term oral opioid therapy in patients with chronic nonmalignant pain. J Pain Symptom Manage 1992;7:69–77.
7. Paice JA, Penn RD, Shott S. Intraspinal morphine for chronic pain: a retrospective, multicenter study. J Pain Symptom Manage 1996;11:71–80.
8. Mystakidou K, Parpa E, Tsilika E, et al. Long-term management of noncancer pain with transdermal therapeutic system-fentanyl. J Pain 2003;4:298–306.
9. Robinson RC, Gatchel RJ, Polatin P, et al. Screening for problematic prescription opioid use. Clin J Pain 2001;17:220–8.
10. Nedeljkovic SS, Wasan A, Jamison RN. Assessment of efficacy of long-term opioid therapy in pain patients with substance abuse potential. Clin J Pain 2002;18:S39–51.
11. Nozaki M, Akera T, Lee CY, Brody TM. The effects of age on the development of tolerance to and physical dependence on morphine in rats. J Pharmacol Exp Ther 1975;192:506–12.
12. Wang Y, Mitchell J, Moriyama K, et al. Age-dependent morphine tolerance development in the rat. Anesth Analg 2005;100:1733–9.
13. Inturrisi CE, Foley KM. Narcotic analgesics in the management of pain. In: Kuhar M, Pasternak G, eds. Analgesics: neurochemical, behavioral, and clinical perspectives. New York: Raven Press; 1984:257–88.
14. Cherny NI. Opioid analgesics: comparative features and prescribing guidelines. Drugs 1996;51:713–37.
15. Donner B, Zenz M, Tryba M, Strumpf M. Direct conversion from oral morphine to transdermal fentanyl: a multicenter study in patients with cancer pain. Pain 1996;64:527–34.
16. Pereira J, Lawlor P, Vigano A, et al. Equianalgesic dose ratios for opioids: a critical review and proposals for long-term dosing. J Pain Symptom Manage 2001;22:672–87.
17. Moulin D, Iezzi A, Amireh R, et al. Randomised trial of oral morphine for chronic non-cancer pain. Lancet 1996;347:143–7.
18. Tennant FS, Uelmen GF. Narcotic maintenance for chronic pain: medical and legal guidelines. Postgrad Med 1983;73:81–94.
19. Neil A, Kayser V, Chen YL, Guilbaud G. Repeated low doses of morphine do not induce tolerance but increase the opioid antinociceptive effect in rats with a peripheral neuropathy. Brain Res 1990;522:140–3.
20. Backonja MM, Miletic G, Miletic V. The effect of continuous morphine analgesia on chronic thermal hyperalgesia due to sciatic constriction injury in rats. Neurosci Lett 1995;196:61–4.
21. Rowbotham MC, Twilling L, Davies PS, et al. Oral opioid therapy for chronic peripheral and central neuropathic pain. N Engl J Med 2003;348:1223–32.
22. Ballantyne JC, Mao J. Opioid therapy for chronic pain. N Engl J Med 2003;349:1943–53.
23. Mercadante S. Opioid rotation for cancer pain: rationale and clinical aspects. Cancer 1999;86:1856–66.