Access to the PPM Journal and newsletters is FREE for clinicians.
10 Articles in Volume 6, Issue #3
A Muscular Approach to Headache
Adjuvant Analgesia for Management of Chronic Pain
Breakthrough Pain In Non-Cancer Patients
Case Presentation of Munchausen Syndrome
Electroanalgesic Medical Device
On Knowing
Opioid Malabsorption: Can You Stomach This?
Sedation Safety and Comfort
The American Board of Independent Medical Examiners (ABIME)
The Role of MMPI-2 in Assessment of Chronic Pain

Breakthrough Pain In Non-Cancer Patients

An observational study of breakthrough pain in non-cancer patients with chronic low back pain

Breakthrough pain (BTP) has been defined as a transitory exacerbation of pain that occurs on a background of otherwise stable pain in cancer patients receiving chronic opioid therapy.1 Additional opioids, such as short-acting breakthrough narcotic medications, with half-lives typically lasting three to four hours, are commonly advocated for BTP in cancer patients with a shortened life expectancy.2,3 While several studies have examined, in depth, the phenomenon of BTP that is related to malignancy in cancer patients,1,4-7 the same has not yet been accomplished for non-cancer chronic pain patients. Patients treated with maintenance doses of opioids for chronic non-cancerous pain also experience transitory flares on a background of otherwise stable pain. To date, characteristics of non-cancerous BTP have been assumed to be the same as cancer-related BTP.

A limited number of studies have begun to examine BTP in the non-cancerous population. The objective of this study was to describe characteristics of BTP in a non-cancerous group of patients with low back pain refractory to one or more surgical interventions. Thirty patients, whose pain was controlled by pharmacological methods that included a long-acting maintenance dose of opioid analgesic for chronic pain supplemented with other medications (both opioid and non-opioid) for breakthrough episodes, were enrolled in this study.

Table 1: Demographics
    N (% or ±SD)
Sex Male 17 (56.7)
  Female 13 (43.3)
Race Caucasian 27 (90.0)
  African American 2 (6.7)

Hispanic 1 (3.3)


  50 (±12) years
Body Mass Index   29.8 (±7.1)
Employment Status Disabled 18 (60.0)

Retired or unemployed 7 (23.3)

Employed (full- or part-time) 5 (16.7)
Number of surgical attempts to correct low back pain 2.7 (Range 1-5)
Average per day dose of long-acting opioid (N=28*) 261 (±325) mg morphine equivalents
Average per day dose of short-acting opioid (N=15) 76 (±78) mg morphine equivalents
*Two patients did not record the dose of their long-acting opioid.


The study was approved by the Washington University Human Studies Committee and utilized a waiver of consent. This pilot study used a survey to evaluate a patient’s perception of chronic and breakthrough pain experienced over the previous 30 days. Eligible patients attending the author’s pain management practice from March 2004 to July 2004 were invited to participate. Enrolled patients had chronic non-cancerous low back pain for a minimum of one year, failed back surgery, used a long-acting maintenance dose of opioid that patients subjectively reported provided at least a 50% reduction in pain, and experienced intermittent episodes of BTP. The patient’s pain management had been optimized using either long acting opioids or around the clock short acting opioids. In some cases the maintenance medication was supplemented with short acting medications (opioids, anti-epileptics, muscle relaxants, antidepressants, and NSAIDS). Patients with cancer, chronic pain not related to the back, or a long-acting opioid dose adjustment in the previous 30 days were excluded from this study.

Validated questionnaires were unavailable for the assessment of BTP in this patient group. Therefore, a questionnaire, adapted from two cancer BTP questionnaires, was designed to elicit information regarding the patients’ perception of both chronic pain and BTP.1,5 Much of the information collected relied on patient recall for a thirty day period prior to administration of the questionnaire. The data collected included: demographic information (see Table 1), descriptive terms of pain (see Table 2), characteristics of BTP (see Table 3), pharmacological and non-pharmacological treatments used for the chronic low back pain (Table 4), and precipitating events of BTP (Table 5).

All analyses were carried out using the software program SPSS 13.0 for Windows (SPSS Inc., Chicago, IL). Both descriptive and inferential statistical methods were employed. All testing was based on determining statistical significance at a two-sided alpha level of 0.05. The study sample was described using measures of central tendency (mean and median) and dispersion (standard deviation and range) for continuous variables and frequency and percentage for categorical variables. Spearman’s rho statistic was used to evaluate the association between continuous/ ordinal scaled variables. Chi-square or Fisher’s Exact tests were used, as appropriate, to compare categorical variables. Mann-Whitney tests were used to compare the distribution of continuous and ordinal scaled variables between 2 categories of categorical variables (see Table 6).


Thirty eligible patients completed the questionnaire. The average chronic baseline pain was 5.5 ± 1.6 on a 0 to 10 scale (10 being the worst pain). The average number of BTP episodes was 6 and the average length of BTP episode was 118 minutes. The most frequent word chosen to describe BTP was sharp, whereas aching was the word chosen most often to describe chronic pain. Seventy-nine percent of the patients in our study indicated that there was a precipitating event 50% or more of the time, and 43% of patients indicated that BTP occurred with end of long-acting opioids dosing. In comparison, Portenoy’s study, characterizing cancer BTP, reported 55% of patients indicated breakthrough incidences had a precipitating event and 30% indicated the BTP episodes were related to end-of-dosing.1 In the current study, patients chose a mean of 4±3 descriptive terms to describe BTP compared to Fine’s study of hospice patients who chose a mean of 1.6 words.5 It was interesting to note, there was a 15% increase in the number of sensory terms and a 45% increase in evaluative terms used by subjects in our study to describe their chronic pain, as opposed to their BTP. This may indicate that patients find describing chronic pain more difficult than describing the breakthrough episodes and view chronic pain more negatively than the breakthrough episodes. The results of the study are presented in Tables 1-6. The statistical analysis of the data indicated a moderately strong positive correlation between the patient’s average pain rating and the long-acting opioid dose. All other analyses demonstrated no significant differences (see Table 6).

Table 2: Comparison of Descriptor Terms for Chronic
and Breakthrough Pain
    Chronic Ranking* Breakthrough Ranking*
Sensory Descriptors
                  N (%)   N (%)    
  Aching 21 (70.0) 1 12 (40.0) 3
  Sharp 17 (56.7) 2 20 (66.7) 1
  Shooting 16 (53.3) 3 14 (46.7) 2
  Stabbing 16 (53.3) 3 11 (36.7) 4
  Burning 12 (40.0) 4 11 (36.7) 4
  Penetrating 9 (30.0) 5 11 (36.7) 4
  Numb 9 (30.0) 5 4 (13.3) 7
  Throbbing 8 (26.7) 6 9 (30.0) 5
  Tender 6 (20.0) 7 5 (16.7) 6
  Gnawing 5 (16.7) 8 4 (13.3) 7
Affective/Evaluative Descriptors
  N (%)   N (%)    
  Nagging 18 (60.0) 1 8 (26.7) 1
  Tiring 13 (43.3) 2 6 (20.0) 3
  Miserable 12 (40.0) 3 7 (23.3) 2

9 (30.0) 4 5 (16.7) 4
  Unbearable 8 (26.7) 5 7 (23.3) 2
* Ranking done by % of subjects using the term
Table 3: Pain Characteristics
Characteristic Chronic* BTP*
Average pain rating 5.5 (±1.6) 6.0 (±1.0)
Worst pain rating 8.1 (±1.8) 9.0 (±1.0)
Mean number of words used to describe pain 6 (±3) 4 (±3)
Number of sensory words used to describe pain 119 101
Number of evaluative terms used to describe pain 60 28
Mean Number of BTP episodes NA 6 (±5)
Median Number of BTP episodes NA 3
Average duration of episodes (minutes) NA 118 (±105)
*N (±SD)    
Table 4: Pain Management Therapies Reported by Patients
Pain Therapy N (%)*
Long-Acting Opioid Treatments (Total per day dose range)t
Fentanyl (25-2000 mcg) 9 (30.0)
OxyContin (40-640 mg) 8 (26.7)
Methadone (15-110 mg) 8 (26.7)
Morphine (29-1500 mg) 6 (20.0)
Short-Acting Opioid Treatments (Total per day dose range)t
Fentanyl (200-400 mcg) 6 (20.0)
Hydrocodone (15-60 mg) 5 (16.7)
Oxycodone (10-40 mg) 3 (10.0)
Hydromorphone (12-16 mg) 2 (6.7)
Morphine (90-180 mg) 2 (6.7)
Codeine (120 mg) 1 (3.3)
Non-Opioid Pharmaceuticalst
Acetaminophen 5 (16.7)
Non-Steroidal Anti-Inflammatory Drugs General 4 (13.3)
Cyclooxygenases-2 Inhibitor 7 (23.3)
Muscle Relaxant 7 (23.3)
Anti-Epileptic 5 (16.6)
Anti-Depressant 3 (10.0)
Non-Pharmaceutical Therapiestt
Spinal Cord Stimulator 2 (6.7)
Physical Therapy 28 (93.3)
Chiropractor 15 (50.0)
Massage 15 (50.0)
Acupuncture 9 (30.0)
Cognitive Therapy 12 (40.0)
* Number and percentage of patients reporting the use of each treatment
t Patient report of current usage of long-acting medication
tt Patient report of past treatment
Table 5: Precipitation of Breakthrough Pain
Event N (%)*
Related to the dosing of the long-acting opioid (N=28)* 12 (42.9)
Precipitating events (N=29)**
Some of the time 6 (20.7)
50% of the time 12 (41.4)
100% of the time 11 (37.9)
Occurred spontaneously (N=30)
Never 3 (10.0)
Some of the time 10 (33.3)
50% of the time 6 (20.0)
Most of the time 8 (28.6)
100% of the time 3 (10.0)
Occurred with movement (N=30)
Some of the time 5 (16.7)
50% of the time 6 (20.0)
Most of the time 8 (26.6)
100% of the time 11 (36.6)
* Number and percentage of patients reporting the use of each treatment
** N<30 represents missing data on patient questionnaire


Since BTP has, to date, primarily only been described in cancer populations, a comparison of BTP in a cancer and a non-cancer population is presented even though it may not be applicable. Substantial differences exist in the basic population of this study versus the cancer studies. First, this patient population had similar pathologies and etiologies, namely low back pain pathology that did not respond to surgery. In the cancer studies, the location and etiologies of the pain varied. Second, the treatment of pain in these two populations is different. Cancer patients frequently have a shortened life expectancy and so opioids are often used more aggressively with the endpoint defined by analgesic side effects or intolerance. The shortened life expectancy is usually not a factor for patients with non-cancerous chronic pain and so the approach to non-cancerous pain management focuses on a broader range of therapeutics with less reliance on opioids. Concerns of inevitable tolerance often results in less aggressive use of opioids in non-cancerous chronic pain states and the increased use of other treatments such as antidepressants and anti-epileptics. Furthermore, the progressive nature of cancer potentially predisposes such patients to a progressive increase in their pain states. Chronic low back pain does not normally demonstrate the same type of progressive increase in pain. Finally, the cancer populations studied were predominately inpatients whereas the current study involved outpatients.

Despite these population differences, similarities in the characterization of BTP were noted. The average number per day of BTP episodes in our low-back pain population was 6 compared to 4, 3 and 7 episodes per day reported in three different cancer studies.1,5,7

Table 6: Statistical Correlations
Chronic Pain Correlations P
1. Long-acting opioid dose vs average chronic pain rating .022
2. Number of words used to describe chronic pain vs average chronic pain rating .150
3. Number of back surgeries vs average chronic pain rating .134
4. Number of chronic pain descriptors vs the number of BTP episodes .094
5. BMI vs average chronic pain rating .983
Number of Breakthrough Pain Episode Correlations
1. Long-acting opioid dose to number of BTP episodes .898
2. Number of back surgeries to number of BTP episodes .501
3. Short-acting opioids dose to number of BTP episodes .955
4. BMI and number of BTP episodes .104
BTP Rating Correlations
1. Long-acting opioid dose to worst BTP rating .295
2. Number of words used to describe BTP to worst BTP rating .387
3. Number of back surgeries to worst BTP rating .086
4. BMI and worst BTP rating .158

Although the number of BTP episodes reported by patients was similar between the non-cancer and the cancer studies, the duration of the episodes appears to be longer in the non-cancer patients. The average duration of the BTP (118 minutes) in this non-cancer population was higher than the duration reported in two cancer studies (30 minutes and 52 minutes).1,5 One additional cancer study reported that 96% of the patients had a duration of pain less than 60 minutes, with the largest number of patients (74%) reporting BTP lasting 30 minutes or less.7 These results may be reflective of differences in the pathology between cancerous and non-cancerous pain, mobility of inpatients and outpatients, or aggressiveness of the treatment regimens between cancer and non-cancer patients.

There are serious limitations to this study. The subjects were asked to complete a questionnaire that required recall from the previous 30 days. Retrospective studies have an inherent flaw in relying on patient recall.8 In particular, asking a subject to recall the average number of BTP episodes per day and the length of these episodes allows for a large margin of error. Further study bias is added when there is missing data due to a lack of specific controls during the period for which the data was created. The survey instrument used in this study was not independently validated and was administered by the patient’s clinician and thus potentially introduced observer bias. No information was collected about psychosocial issues that may affect the patient’s view of chronic pain, BTP, and the use of medications and substances of abuse.

Regardless of these study limitations, this pilot study provides a basic foundation for future, more systematic prospective studies to evaluate the characteristics of breakthrough pain in non-cancerous disease states. Defining optimized chronic pain management is one of the key points to be determined. In this study, three of the patients reported 13 or more breakthrough episodes a day, despite reporting a 50% reduction in their chronic pain. This raises the question as to whether these patients’ chronic pain was being optimally managed. In addition, a future study should include an evaluation of the patient’s substance abuse and psychosocial issues. These issues, which can be associated with opioid usage, should be examined in relationship to optimized pain management, as well as the number and length of breakthrough episodes. Finally, BTP should be examined in a variety of non-cancer pain populations. The current study focused on a patient population with chronic low back pain as a result of failed back surgery. These patients had recalcitrant pain problems and had been referred to a tertiary pain management group, therefore their perception of chronic and BTP may not be representative of other non-cancer chronic pain populations.


This study indicates that the frequency of BTP in this group of non-cancer chronic low back pain patients is similar to that for patients diagnosed with cancer. Patients in both populations described BTP in similar terms and described similar precipitating events, such as movement and end-of-dose of long-acting pain medication. However, the duration of breakthrough episodes was reported as being longer in this non-cancer patient population. Further studies need to be conducted to validate and refine the results observed in this limited study. BTP occurs with intractable pain regardless of the underlying etiology. The authors believe that a better understanding of non-cancer BTP will enable physicians to more effectively assess and treat BTP in the context of optimally managing chronic non-cancer pain.

Last updated on: January 10, 2012
close X