Chronic Pain in Persons With Neuromuscular Disease

Participants 

The study participants were 193 respondents to a survey on the nature and scope of pain in persons with NMD. Subjects for this study were recruited from 2 sites: (1) an NMD rehabilitation clinic at the University of Washington (UW) Medical Center and (2) an NMD Clinic and an NMD research and training center at the Department of Physical Medicine and Rehabilitation, University of California, Davis (UCD). Inclusion criteria for this study were primary diagnosis of NMD and chronologic age of 18 or older.

Of 103 potential subjects approached at the UW NMD clinic, 89 (86%) agreed to participate in the survey study. Most of the potential subjects who declined participation did so because they either lacked interest or were so severely disabled that they did not want to take on another task. Of 40 potential subjects approached at the UCD NMD clinic, 36 (90%) agreed to participate in the survey study. Two hundred sixty-six surveys were also mailed to people on the UCD NMD research and training center patient list. Of these, 22 were returned because the patient had moved or the address was otherwise wrong. Sixty-eight surveys were returned (response rate, 29%), for a total of 104 surveys from UCD (36 from the clinic, 68 from the mail survey). Basic demographic and descriptive information for the 193 survey respondents are listed in table 1. Diagnostic information of the study participants (type of NMD) is presented in table 2.

Table 1. Demographic Characteristics of Survey Respondents

	Variable	UW Sample	UCD Sample	Combined Sample
	N	89	104	193
	Sex (%) (M/F)	54/46	42/58	48/52
	Age (y)
	Mean ± SD	51.7±15.7	51.8±15.7	51.7±15.6
	Range	19–84	18–83	18–84
	Ethnic group (%)
	White	84	89	87
	African American	1	1	1
	Asian	6	1	3
	Hispanic/Chicano	3	4	4
	Native American	0	2	1
	Other	6	4	5

Abbreviations: F, female; M, male; SD, standard deviation.

Table 2. NMD of Survey Respondents

	Diagnosis	N	%
	LGMD	44	23
	CMT	35	18
	ALS	30	16
	MMD	26	14
	FSHMD	18	9
	PPS	13	7
	Other⁎	27	14

Abbreviation: PPS, postpoliomyelitis syndrome.

⁎ Includes all survey participants who had a diagnosis that was infrequent (n<10) in the sample, including patients with SMA (n=8), primary lateral sclerosis (n=3), Duchenne’s muscular dystrophy (n=2), Becker’s muscular dystrophy (n=1), CMT derivative (n=1), Emery-Dreyfus (n=1), cerebellar degeneration (n=1), mitochondrial myopathy (n=1), congenital myopathy (n=1), facioscapulohumeral dystrophy (n=1), benign focal amyotrophe (n=1), myotonia congenita (n=1), and patients attending one of the NMD clinics with clear NMD-like symptoms but who do not yet have a specific NMD diagnosis (n=5).

Procedures 

Measures 

All participants were asked to provide basic demographic and descriptive information and to indicate whether they had experienced any pain in the past 3 months (other than occasional headaches or menstrual cramps). Those who responded affirmatively to the latter question were then asked to complete measures of average and worst pain intensity, the quality of pain experience, pain interference with functioning, the location(s) of pain, general QOL, treatments received for pain, and the relief provided by those treatments. The specific items and questionnaires used to assess these domains are described later. The survey took 45 to 60 minutes to complete for those with pain and about 10 minutes for those without pain.

Data Analysis 

Descriptive statistics (rates and percentages of participants reporting pain, means and standard deviations [SDs] of pain intensity ratings, rates and percentages of participants reporting moderate and severe pain) for each NMD diagnostic group, and for the sample as a whole, were computed to describe the frequency and severity of pain in the sample. Means and SDs of ratings of specific pain qualities, and the impact of pain on specific activities, were computed and then compared by using a repeated-measures analysis of variance (ANOVA), followed by univariate tests to help explain any significant differences found in pain quality and impact. The percentage of respondents with pain who reported pain in 11 specific body sites were computed for each diagnostic group and for the study sample as a whole to help describe the location(s) of pain in the sample. The means and SDs of the SF-36 scale scores were then computed for each diagnostic group and for the sample as a whole, to describe general QOL among the respondents. We performed t tests to compare the SF-36 scale scores of the current sample with those from a US normative sample. In addition, an omnibus ANOVA was performed, to determine whether there were any differences between the diagnostic groups on the SF-36 scales; subsequent univariate tests were conducted to help interpret any significant overall group effects found. Finally, the percentages of respondents who reported using different pain treatments (among all the respondents with pain, as well as a subgroup of respondents reporting severe pain) were computed, as well as the means and SDs of the amount of relief obtained with each pain treatment, to help describe the pain treatments used by the participants.

Frequency and Severity of Pain 

Table 3 lists the percentage of the sample that reported experiencing pain in the past 3 months (73%), as well as the percentage within each diagnostic group that reported pain. As can be seen, about three quarters of the sample reported experiencing pain, with the percentage ranging from 60% (for patients with ALS) to 100% (for patients with postpoliomyelitis syndrome [PPS]). Average pain, for those who reported having pain, was in the low to moderate range (across diagnostic groups, 4.00–6.28; overall, 5.06; see table 3), and worst pain was in the high to moderate range (5.59–7.33; overall, 6.65), but there was considerable variability in reported average and worst pain intensity within each diagnostic group, as indicated by the relatively large SDs associated with both pain intensity ratings. About one quarter (27%) of the sample reported that the average pain intensity they experienced was severe (range from 9% for “other” NMD diagnosis to 50% for patients with MMD), and a little more than half (56%) reported that their worst pain was in the severe range.

Table 3. Pain Frequency and Severity for All Subjects and Across NMD Diagnostic Groups

	Pain Measure	All Subjects	NMD Diagnosis
			LGMD	CMT	ALS	MMD	FSHMD	PPS	Other
	Response to question concerning the presence or absence of pain in the past 3 months
	% reporting pain	73	64	75	60	69	89	100	82
			Response to 0–10 rating scale of “average pain in the past week”
	Average pain±SD⁎	5.06±2.58	5.25±2.52	5.15±2.57	5.28±2.61	6.28±3.21	4.25±2.24	5.31±2.53	4.00±2.00
	% reporting moderate average pain†	28	39	27	33	11	19	15	36
	% reporting severe average pain‡	27	25	27	28	50	19	39	9
			Response to 0–10 rating scale of “worst pain in the past week”
	Worst pain±SD⁎	6.65±2.60	6.89±2.63	6.69±2.48	7.06±2.31	7.33±2.87	6.06±2.89	7.08±2.43	5.59±2.54
	% reporting moderate worst pain†	18	18	19	28	0	25	8	23
	% reporting severe worst pain‡	56	61	54	61	72	44	69	36

⁎ Assessed by using a 0 (no pain) to 10 (pain as bad as can be) scale for those subjects with pain. † Moderate pain is pain rated as 5 or 6 on a 0 to 10 scale. ‡ Severe pain is pain rated as 7 or greater on a 0 to 10 scale.

Pain Quality 

The results concerning the quality of NMD-related pain reported by the study participants are presented in table 4. There was clear variation in the description of NMD pain as rated by the study participants, with a significant (F=46.07, P<.001) difference across descriptors as tested by repeated-measures ANOVA. The results of univariate paired t tests indicate that “cold” and “itchy” were rated as significantly (P<.05) lower than all the others. On the other hand, “deep,” “tiring,” “sharp,” and “dull” descriptors were frequently used and were rated as significantly higher than all the other pain descriptors.

Table 4. Pain Quality Ratings Among Subjects With Pain

	Pain Quality Item⁎	Mean±SD
	Sharp	5.77±3.46a
	Hot	3.81±3.44bc
	Dull	5.65±2.84a
	Cold	1.91±3.18d
	Sensitive	2.92±3.36ef
	Itchy	1.96±2.95g
	Tiring	6.18±3.28ag
	Sickening	2.79±3.39e
	Fearful	3.30±3.42bf
	Punishing	4.14±3.69c
	Unpleasant	6.07±2.48g
	Deep	6.70±2.57h
	Surface	4.11±2.85c

NOTE. Different superscripts associated with the mean ratings for each pain descriptor indicate significant (P<.05) differences between descriptors. For example, the superscript “a” next to the mean associated with sharp pain (5.77) indicates that this mean is statistically significantly larger than the means associated with hot (3.81), cold (1.91), sensitive (2.92), itchy (1.96), sickening (2.97), fearful (3.30), punishing (4.14), and surface (4.11) pain (none of these have a superscript “a” and they are smaller means), statistically significantly smaller than the means associated with unpleasant (6.07) and deep (6.70) pain (neither of these has a superscript “a” and they are larger means), and not statistically significantly different from the means associated with dull (5.65) and tiring (6.18) pain (both of these have a superscript “a”).

⁎ All rated on a 0 (eg, not sharp, hot, dull) to 10 (eg, the most sharp, hot, dull sensation imaginable) scale.

No clear pattern of pain description appears to differentiate pain associated with any 1 NMD diagnosis from the others. Only 1 omnibus test for differences between diagnostic groups was statistically significant; this was for ratings of “fearful” pain (F=2.46, P<.05). Univariate analyses indicate that much of this effect was because of the fact that patients with ALS (mean ± SD, 5.17±4.12) and MMD (mean, 4.83±3.73) reported their pain as significantly more fearful than patients with CMT (mean, 2.15±2.52), FSHMD (mean, 2.13±2.85), and “other” NMD diagnosis (mean, 2.26±3.49).

Pain Interference 

Table 5 lists the average ratings of pain interference with the 10 activities of the modified BPI pain interference scale. An omnibus ANOVA of the interference scores across activities was significant (F=12.76, P<.001), which suggests some variability in the impact of NMD pain on different activities. Univariate paired t tests between the ratings indicated that this effect was because of the fact that pain interfered less with relations with other people and self-care (P<.05) than with the other activities. Also, interference with normal work (including work outside the home and housework) and with recreational activities was significantly higher than pain interference with general activities, except interference with sleep.

Table 5. Pain Interference Among Subjects With Pain

	Pain Interference Item⁎	Mean±SD
	General activity	3.70±3.10ab
	Mood	3.81±2.89ab
	Mobility (ability to get around)	3.99±3.34b
	Normal work	4.46±3.39c
	Relations with other people	2.56±2.85d
	Sleep	4.04±3.48abc
	Enjoyment of life	3.91±3.07b
	Self-care	2.60±3.24d
	Recreational activities	4.63±3.59c
	Social activities	3.41±3.21a

NOTE. Different superscripts associated with the mean ratings for each activity indicate significant (P<.05) differences between activity items. For example, the superscript “b” next to the mean associated with impact of pain on enjoyment of life (3.91) indicates that this mean is statistically significantly larger than the means associated with the impact of pain on relations with other people (2.56), self-care (2.60), and social activities (3.41) (these means are smaller and do not have a superscript “b”), statistically significantly smaller than the means associated with the impact of pain on normal work (4.46) and recreational activities (4.63) (these means are larger and do not have a superscript “b”), and not statistically significantly different from the means associated with the impact of pain on general activity (3.70), mood (3.81), and sleep (4.04) (each of these means has a superscript “b”).

⁎ All rated on a 0 (does not interfere) to 10 (completely interferes) scale.

Analyses were also performed to examine differences in the extent to which pain interferes with daily activities between the diagnostic groups. The results of omnibus repeated-measures ANOVAs yielded significant (P<.05) effects for the diagnostic group for 4 of the activities: general activity, relations with other people, self-care, and social activities. Univariate analyses indicated that, for the general activity item, patients with ALS (mean, 5.00±3.68) and MMD (mean, 5.00±3.40) report significantly greater pain interference than patients with LGMD (mean, 2.57±2.71) or patients with an “other” NMD (mean, 2.29±2.17) diagnosis. Similarly, on the relations with other people item, patients with ALS (mean, 3.72±3.48) and MMD (mean, 4.33±3.73) reported significantly greater pain interference than patients with a CMT (mean, 1.38±2.14) or “other” NMD diagnosis (mean, 1.71±1.45). For self-care, the significant effects were due to the fact that patients with a CMT (mean, 1.08±2.20) and “other” NMD diagnosis (mean, 1.52±2.11) reported significant lower levels of pain interference on this activity than patients with ALS (mean, 4.75±3.86). Finally, for social activities, patients with CMT (2.08±2.64) reported significantly less pain interference than patients with ALS (mean, 4.88±4.20), MMD (mean, 4.39±3.11), and PPS (mean, 4.31±3.09).

Pain Location 

The percentages of participants with pain who reported pain at each location are presented in table 6. The most frequent pain site, overall, was back (49%), followed by leg (47%), shoulder (43%), neck (40%), buttock and hip(s) (37%), feet (36%), arm(s) (36%), and hand(s) (35%). Head and face (20%), chest (17%), and abdomen and pelvis (16%) pain were relatively rare. Although the rates of pain at different sites were often similar across some NMD diagnoses, there were some interesting exceptions. Patients with CMT, for example, reported rates of pain at different sites that tended to be either very similar to, or lower than, the rates of other diagnostic groups, on average, except when it came to leg (60%) and foot (66%) pain, which were the most frequent pain problems in persons with CMT. Patients with PPS, on the other hand, reported shoulder pain as their most common pain complaint (85%), followed closely by back (77%), hand, neck, foot, and buttock pain (69% each). Back (78%), shoulder (67%), and buttock (61%) pain were the most common among participants with FSHMD as well. There were no locations of pain in participants with other diagnoses (LGMD, ALS, MMD, “other” NMD) that occurred with a frequency that was greater than 60%.

Table 6. Pain Locations Among Subjects With Pain

	Location	% With Pain
	Head/face	20
	Neck	40
	Shoulder(s)	43
	Back	49
	Chest	17
	Arm(s)	36
	Hand(s)	35
	Buttock/hip(s)	37
	Abdomen/pelvis	16
	Legs	47
	Feet	36

Quality of Life 

The means and SDs of the SF-36 scale scores for all study participants with pain are listed in table 7, along with the means and SDs of the US norms for these scales. As might be expected, the study participants reported significantly greater dysfunction than subjects in the SF-36 normative sample, as measured by the physical functioning (t=16.84, P<.001), role-physical (t=6.29, P<.001), bodily pain (t=13.48, P<.001), general health (t=7.46, P<.001), vitality (t=7.13, P<.001), and social functioning (t=8.20, P<.001) scales. Interestingly, however, no significant differences existed between the study participants and the US norms on the SF-36 role-emotional or mental health scales. The only significant difference to emerge using ANOVAs to test between-group differences was for the SF-36 bodily pain scale (F=2.95, P<.05). The results of the univariate analyses indicate that this effect was primarily due to the significant differences between patients with ALS (mean, 43.84±18.54) and MMD (mean, 44.21±16.78) and patients with CMT (mean, 52.83±12.35) and “other” NMD diagnoses (mean, 57.30±12.07); patients with the former diagnoses had significantly lower scores (indicating greater pain severity on the SF-36 bodily pain scale) than patients with the latter diagnoses.

Table 7. General US and Study Sample (subjects with pain) Norms of SF-36 Scores

	SF-36 Scale Score⁎	General US Norms5	Study Sample
	Physical functioning	84.15±23.28a	50.41±15.98b
	Role-physical	80.96±34.00a	62.28±17.77b
	Bodily pain	75.15±20.34a	51.50±15.74b
	General health	71.95±20.34a	58.76±20.37b
	Vitality	60.86±20.96a	47.95±17.10b
	Social functioning	83.28±22.69a	67.12±20.93b
	Role-emotional	81.26±33.04	79.01±21.44
	Mental health	74.74±18.05	73.23±14.94

NOTE. Values are mean ± SD. Different superscripts associated with the mean ratings for SF-36 scale score indicate significant (all P< .001) differences between general US norms and the study participants for these scales.

⁎ All standardized to range from 0 to 100, with 100 indicating highest level of self-reported functioning.

Pain Treatments and Treatment-Related Pain Relief 

Table 8 lists the percentage of patients with pain who had tried each pain treatment at least once previously as well as the percentage of those who are currently using the treatment and the average amount of pain relief produced by the treatment. Because it is likely that many patients with only mild or even moderate pain may not seek pain treatments, the percentages of patients with severe pain (average pain ≥7 on a 0–10 scale) who have tried and currently use the pain treatments listed are also presented in table 8. As can be seen, ibuprofen and aspirin (61% of all respondents with pain, 47% of those with severe pain) are the medications most often tried for pain management, followed by acetaminophen (47% all subjects, 37% with severe pain), physical therapy (43%, 50%), and narcotic analgesics (35%, 42%).

Table 8. Percentage of Participants With Pain Who Have Tried Each Treatment, Percentage of Participants Who Still Use the Treatment, and Average Relief Rating Associated With Each Treatment

	Pain treatment	All Subjects With Pain (n=141)		Subjects With Severe Pain (n=38)
		% Tried/% Still Use⁎	Average Relief±SD†	% Tried/% Still Use⁎	Average Relief±SD†
	Ibuprofen, aspirin	61/65	5.22±2.83	47/78	4.25±3.05
	Acetaminophen	47/58	4.11±2.93	37/50	3.31±2.87
	Physical therapy	43/42	4.54±2.66	50/37	3.89±2.25
	Narcotics	35/63	6.37±2.74	42/56	5.75±2.79
	Massage	34/44	5.48±2.73	34/39	4.91±3.36
	Neurontin	18/50	4.78±3.02	24/56	4.57±3.05
	Muscle relaxants	18/60	5.78±2.88	21/50	4.25±1.26
	Tricyclic antidepressants	15/38	4.53±3.28	18/29	5.43±2.99
	Acupuncture	11/25	5.29±3.22	3/100	6.00±0.00
	Magnets	11/25	3.13±3.16	11/25	1.75±2.87
	Biofeedback/relaxation training	8/55	4.42±2.50	11/75	4.50±1.91
	Counseling	9/67	4.70±2.50	18/71	4.17±2.64
	Chiropractic manipulation	4/85	7.33±3.78	0/NA	ND
	Carbamazepine	4/17	3.80±4.38	8/33	6.33±3.79
	Nerve blocks	3/0	6.75±4.76	5/0	10.00±0.0
	Hypnosis	2/3	5.00±4.24	0/NA	ND

Abbreviation: NA, not applicable; ND, no data.

⁎ “Percentage still use” is the percentage of subjects who reported a history of using the pain treatment at one time who are currently using or receiving the treatment. † Subjects who had used the treatment rated the amount of relief provided by each on a 0 (no relief) to 10 (complete relief) scale.

However, the treatments that provided the greatest pain relief were not necessarily those that are most frequently used. The average relief rating, on a 0 to 10 scale, for chiropractic manipulation was 7.33 for the very few patients (4%) with pain who tried this treatment. Most of these patients reported that they were still receiving this treatment. No patients with severe pain reported ever having tried chiropractic care. Nerve blocks were reported as providing the next highest degree of relief among all the treatments (average relief rating, 6.75), although none of the patients who received these in the past were still receiving nerve blocks. Opioid analgesics were also listed as providing more relief than other pain treatments (average rating, 6.37) and were tried by about a third of the participants with pain overall. Interestingly, however, only about two thirds of those participants who tried opioids for pain were still using this treatment at the time of the survey. Other treatments that provided some pain relief, on average (relief rating ≥5.00 on the 0–10 scale), were ibuprofen and aspirin, massage, muscle relaxants, acupuncture, and hypnosis. Treatments that appeared to provide relatively little relief (relief rating, <4.00) were carbamazepine and magnets, although the former received a relatively high relief rating among those participants with severe pain who tried it (6.33). Across all treatments, there was a fair amount of variability (SD range, 2.50–4.76 for all respondents with pain) in the relief provided by the pain treatments.

Pain Frequency and Severity 

The results indicate that pain is most common in patients with PPS (100% of survey respondents) and perhaps least common in persons with ALS (60%) but still quite common in the latter group. About 25% of the sample reported that they experience severe pain on average. However, there was considerable variability in the frequency of severe pain between diagnostic groups. Severe pain appears to be more common among patients with MMD (50% report severe pain) and PPS (39%) and less common among patients with other NMD diagnoses. Future research is needed to determine underlying mechanisms that may explain the observed differences in the relative frequency of severe pain in persons with NMDs.

Pain Quality 

NMD-related pain was primarily described as tiring, unpleasant, deep, and both sharp and dull by the sample overall. It is not particularly cold, sensitive, or itchy. In addition, the quality of pain did not appear to differ substantially between diagnostic conditions. In general, and with the exception of the descriptor “fearful,” there appeared to be as much variability within as there was between diagnostic groups. To the extent that pain descriptors reflect different underlying biologic causes of pain, these findings suggest that there may be a variety of causes or sources of pain within patients with the same diagnosis, including, perhaps, mechanisms related to muscle cramping, musculoskeletal problems associated with over- or underuse because of mobility restrictions, and nerve damage (causing painful neuropathies). To the extent that future research is able to identify specific treatments that impact specific pain qualities,8 treatments that impact tiring, deep, and/or sharp pain may be particularly effective for persons with NMD-related pain problems.

Pain Interference 

Noteworthy differences between diagnostic groups were found in the extent to which pain interfered with function. For example, pain appeared to have relatively little impact on the study participants’ self-care or relations with other people, on average. On the other hand, pain did appear to have a greater impact on mobility, normal work (including housework), enjoyment of life, and recreational activities compared with other activities assessed by the modified BPI pain interference scale. Each of these activities affected by pain represents important domains of QOL and underscores the necessity of providing appropriate pain treatment to ensure maximum QOL in persons with NMD-related pain.

Although no significant differences were found between the diagnostic groups across all interference items, there were significant effects for general activity, relations with other people, self-care activities, and social activities. Moreover, the patterns of these effects painted a similar picture across the activity domains; when differences were found, patients with ALS and MMD tended to report greater interference with function than did patients with CMT. These findings, if replicated in future studies, suggest that patients with ALS or MMD and pain may be particularly vulnerable to the effects of pain.

Pain Location 

Although the participants in this study reported pain in virtually all possible locations, NMD-related pain appears to occur most frequently in the back, legs, shoulder(s), and neck. It appears relatively less often in the chest and abdomen and pelvis. The pain in the low back and legs may be because of the profound gait abnormalities often associated with NMD.21, 22, 23, 24 Typically, these patients will have a Trendelenburg gait, with increased lumbar lordosis. These diseases will often create significant shoulder girdle weakness and instability as well, which could explain the upper back and neck and shoulder pain.

Although the frequency of pain sites associated with each NMD diagnosis was generally consistent with the frequency of pain sites in the sample overall, there were some interesting differences between some diagnostic groups. For example, the most common pain sites in persons with CMT-related pain were the legs and feet. This is consistent with the nature of CMT, which is a length-dependent neuropathy that affects the legs and feet sooner and more severely than other parts of the body.22

On the other hand, patients with PPS reported shoulder pain as their most common pain complaint (85%), followed closely by back (77%), hand, neck, foot, and buttock pain (69% each). These pain sites would be consistent with an NMD condition that often has widespread and severely debilitating effects on the motor system, creating flaccid extremities, and poorly supported joints and ligaments. Back (78%), shoulder (67%), and buttock (61%) pain were the most common among participants with FSHMD. Back and shoulder pain in patients with FSHMD would be expected, given that these patients typically have severe lumbar lordosis and shoulder girdle instability.25

Quality of Life 

Expected differences between the sample as a whole and US norms on many of the SF-36 scales, and especially those related to physical functioning, were found. However, the most striking finding from the analyses comparing the sample as a whole to the US norms on the SF-36 was the lack of significant difference for the SF-36 mental health and role-emotional scales. These findings suggest that persons with NMD, and even persons with chronic pain and NMD, can adjust to their condition and still report psychologic functioning comparable to persons who are not disabled or have other health problems.

The relatively low levels of functioning assessed by the SF-36 physical functioning, role-functioning, bodily pain, general health, and vitality scales in the study sample compared with US norms does support the need to develop interventions that may help to improve functioning in these areas for persons with NMD-related pain. There is a growing body of research that seeks to develop and test interventions that will improve functioning in persons with NMD. For example, several studies20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31indicate that interventions designed to promote exercise can increase functional capacities in slowly progressive NMDs. In addition, reviews32, 33, 34of the literature have concluded that persons with slowly progressive NMDs should adopt an active lifestyle and proper nutrition as a way of improving QOL and reducing secondary conditions.

Pain Treatments 

The findings concerning pain treatments tried and used, and their relative effectiveness, indicate that a great deal of improvement is needed in the treatment of pain in persons with NMD. First, there are very few treatments that appear to provide significant permanent relief in the study sample. Of the treatments that provided the greatest relief (chiropractic care, nerve blocks), only chiropractic care continued to be used by most patients who had tried it, which suggests that the costs and/or risks of repeated nerve block procedures may not have been worth the benefits, at least among the patients surveyed. The very highly rated relief provided by chiropractic care is intriguing and suggests the strong need for a controlled trial of this treatment to see if it is effective for large numbers of persons with NMD pain and to determine the length of any benefit found.

The most common treatments tried were over-the-counter (OTC) analgesics: ibuprofen, aspirin, and acetaminophen. However, even these were reported to provide only moderate relief (4.11–5.22 on a 0–10 relief rating scale overall; 3.31–4.25 among respondents with severe pain), and not every person with pain who tried these treatments was still using them. Also, even though many of the study participants with severe pain (78%) reported using ibuprofen or aspirin to treat their pain, it must be acknowledged that these patients still reported severe pain.

Other pain treatments had been tried and were still used but by fewer patients. Among the 3 treatments tried by 30% to 50% of the respondents—physical therapy, narcotic analgesics, and massage—only the latter 2 were associated with relief ratings more than 5 on the 0 to 10 relief rating scale. As with the OTC medications, many patients who tried these treatments were no longer using them. For example, only 63% of the patients who tried opioid analgesics (56% of those with severe pain) reported using opioids when they completed the surveys. Clearly, none of these treatments provides effective long-term relief for all the patients.

Other pain treatments were tried even more rarely, including gabapentin (Neurontin), muscle relaxants, tricyclic antidepressants, acupuncture, magnets, biofeedback and relaxation training, counseling, carbamazepine, and hypnosis. Of these, only muscle relaxants, acupuncture, and hypnosis were associated with a relief rating that was 5 or more on the 0 to 10 relief scale in the group overall (although carbamazepine and tricyclic antidepressants were associated with higher relief ratings among those with severe pain who tried these treatments). As a group, these findings suggest that (1) none of the treatments provide a great deal of pain relief for all patients but (2) many pain treatments appear to provide at least some relief for some patients.

The lack of clear efficacy of any 1, or more, of the pain treatments examined here highlights the difficulty of diagnosing and treating pain in persons with NMD. Such pain has a large number of potential causes, including factors related to the disease process itself, as well as (and perhaps more commonly) other factors secondary to the disease process, such as abnormal posture or gait, joint instability or ligament sprain due to severe weakness, muscle cramping, nerve impingement because of muscle atrophy or dystrophy, and prolonged use of a manual wheelchair (eg, overuse). The ideal approach to pain control begins with the identification and then the elimination or treatment of the underlying etiologic factors. An important next step will be to identify the most common causes of pain problems in persons with NMD and then to develop standardized and valid methods for classifying those pain problems. Once such methods are established, research could then identify those pain treatments that are most effective for the different causes or mechanisms of pain in NMD. Such research could help clinicians and their patients move from a trial-and-error to an evidence-based approach for selecting pain treatments, ultimately contributing to a more efficient decrease in the pain and suffering in persons with NMD.

Study Limitations 

Several important limitations of the study should be noted. First, all the data were collected via self-report. Although much of the information—for example, information concerning pain severity and quality—must be collected in this way, it would have been useful to corroborate measures of observable behavior, such as pain treatments tried and disability level, from additional sources, such as medical records and the reports of spouses or others who know the patient well. In addition, although many of the measures and questionnaires used in the survey instrument have shown reliability and validity in previous samples of persons with chronic pain, their psychometric properties in people with NMD have not yet been established. Also, because this was primarily a descriptive study and many variables were examined, there were a large number of statistical tests performed. This increases the chances of type I errors (finding a significant difference by chance). On the other hand, the relatively low number of study participants within each diagnostic group limits the power to detect differences between the diagnostic groups, which is associated with type II errors (not finding a difference between 2 samples that does exist in the population). Thus, the results of the analyses involving comparisons between diagnostic groups should be considered tentative at this time and in need of replication. Finally, the pain treatments listed in the survey included physical therapy, which usually consists of a variety of specific interventions, including both passive (eg, passive range of motion and stretching, massage, application of transcutaneous nerve stimulation, heat, or cold) and active (eg, training and supervision of specific muscle strengthening exercises) treatments. It is possible, even likely, that these different interventions have different effects on pain. Future research examining the frequency and effects of pain treatments should assess each of these different interventions separately and should not merely ask about “physical therapy” in general.