Chiropractic Treatment vs Self-Management in Patients With Acute Chest Pain: A Randomized Controlled Trial of Patients Without Acute Coronary Syndrome

Article Outline

• Abstract
• Methods
  • Settings and Participants
  • Trial Procedures
  • Randomization and Blinding
  • Description of Interventions
    • Chiropractic Treatment Program
    • Self-Management Program
  • Follow-Up Procedures
  • Outcome Measures
  • Sample Size
  • Statistical Analysis
• Results
  • Participants
  • Chiropractic Treatment
  • Prevalence of Pain
  • Numeric Change in Pain Intensity
  • Self-Perceived Change in Chest Pain
  • Quality of Life Scores (SF-36)
  • Analysis of Sensitivity of Results
• Discussion
  • Interpretation of Results
  • Strengths
  • Limitations
  • Lessons Learned and Questions to Be Answered
• Conclusions
• Funding Sources and Potential Conflicts of Interest
• References
• Copyright

Abstract 

Objective

The musculoskeletal system is a common but often overlooked cause of chest pain. The purpose of the present study is to evaluate the relative effectiveness of 2 treatment approaches for acute musculoskeletal chest pain: (1) chiropractic treatment that included spinal manipulation and (2) self-management as an example of minimal intervention.

Methods

In a nonblinded, randomized, controlled trial set at an emergency cardiology department and 4 outpatient chiropractic clinics, 115 consecutive patients with acute chest pain and no clear medical diagnosis at initial presentation were included. After a baseline evaluation, patients with musculoskeletal chest pain were randomized to 4 weeks of chiropractic treatment or self-management, with posttreatment questionnaire follow-up 4 and 12 weeks later. Primary outcome measures were numeric change in pain intensity (11-point box numerical rating scale) and self-perceived change in pain (7-point ordinal scale).

Results

Both groups experienced decreases in pain, self-perceived positive changes, and increases in Medical Outcomes Study Short Form 36-Item Health Survey scores. Observed between-group significant differences were in favor of chiropractic treatment at 4 weeks regarding the primary outcome of self-perceived change in chest pain and at 12 weeks with respect to the primary outcome of numeric change in pain intensity.

Conclusions

To the best of our knowledge, this is the first randomized trial assessing chiropractic treatment vs minimal intervention in patients without acute coronary syndrome but with musculoskeletal chest pain. Results suggest that chiropractic treatment might be useful; but further research in relation to patient selection, standardization of interventions, and identification of potentially active ingredients is needed.

Key Indexing Terms: Chiropractic, Manipulation, Spinal, Chest Pain, Randomized Clinical Trial

Acute chest pain is the hallmark of acute coronary syndrome (ACS) and accounts for 5% to 6% of all admissions to emergency departments in Europe and the United States; however, only 20% to 25% of these admissions turn out to be caused by ACS.¹, ² Between 1990 and 2000, most hospitals observed an increase in the overall number of admissions for suspected ACS, caused primarily by a doubling of patients with angina pectoris or undifferentiated chest pain.³

Patients with undifferentiated chest pain account for approximately 20% of admissions for suspected ACS.², ³, ⁴ Commonly, they leave emergency departments without a definite diagnosis or a plausible explanation for their pain.⁵ Despite thorough diagnostic assessment, many continue to have recurrent episodes of chest pain, leading to anxiety, reduced quality of life, and frequent contacts with the health care system.⁶, ⁷, ⁸, ⁹ In the United States, it has been estimated that it costs approximately $8 billion for the initial care of patients suspected of having ACS, but who are subsequently found not to have coronary artery disease,¹⁰ whereas the long-term economic cost of undiagnosed chest pain is unknown.⁴

In patients with undifferentiated chest pain, musculoskeletal dysfunction may be an overlooked source of pain,¹¹, ¹² for which chiropractic treatment has been suggested as being beneficial in case reports.¹³, ¹⁴ In 2004, our research group developed a standardized evaluation protocol to identify patients with musculoskeletal chest pain among patients with known or suspected stable angina pectoris, and treated them with chiropractic therapy in a nonrandomized clinical trial.¹⁵, ¹⁶ The results suggested that patients did benefit from the chiropractic treatment; but the study did not allow us to fully elucidate the value of chiropractic treatment in this category of patients, nor did it consider patients with acute chest pain.

The purpose of this randomized controlled trial was to evaluate the relative effectiveness of 2 conservative treatment approaches in patients with an acute episode of musculoskeletal chest pain: (1) chiropractic treatment that included spinal manipulation and (2) self-management as an example of minimal intervention. Effectiveness was assessed by questionnaires 4 and 12 weeks after randomization.

Back to Article Outline

Methods 

Settings and Participants 

This study was carried out in an emergency cardiology department at a 1000-bed, urban, university hospital in Denmark and at 4 local chiropractic clinics from 6 August 2006 to 31 March 2008. The study was approved by the regional ethics committee of Vejle and Funen counties, Denmark (approval no. #VF 20060002), and registered at ClinicalTrials.gov (identification no. NCT00462241).

All patients presenting at the emergency unit with an episode of acute chest pain underwent the routine diagnostic procedures performed by specialist cardiology nurses under cardiologist supervision. They covered rapid diagnostic assessment for ACS using electrocardiogram and biochemical cardiac marker testing, that is, creatine kinase MB (mass) levels on admission and 6 to 9 hours later, and troponin T levels at least 6 hours after the worst symptoms. When a patient was discharged from the unit, the study clinician (MJS) screened the patient's records to identify his/her eligibility for our study. Participants should not have had a diagnosis of ACS or another definite cardiac or medical diagnosis, should be aged 18 to 75 years, should have had a primary complaint of acute chest pain for less than 7 days' duration, should be a resident of the local county, and should be able to read and understand Danish. In addition, participants should have undergone diagnostic procedures to rule out ACS and should not have shown significant comorbidity or contraindications for spinal manipulative therapy. Exclusion criteria comprised previous ACS, prior percutaneous coronary intervention or coronary artery bypass grafting, inflammatory joint disease, insulin-dependent diabetes, fibromyalgia, malignant disease, major osseous anomaly, osteoporosis, apoplexy or dementia, inability to cooperate, and pregnancy. In each case, the cause for exclusion was noted.

Trial Procedures 

After providing written informed consent, the participants were assessed at baseline by the study clinician using a standardized and previously validated study protocol.¹⁵, ¹⁷ The protocol involves a case history and a clinical health examination, including manual examination of the spine and chest wall, to diagnose possible musculoskeletal chest pain. Demographic and clinical information was collected through patient self-report questionnaires and checklists used by the study clinician. Detailed trial procedures are described elsewhere.¹⁸

Randomization and Blinding 

Only patients with a positive diagnosis of musculoskeletal chest pain were eligible for randomization. The randomization schedule was computer generated by a researcher not involved in the study and concealed from the study team. Consecutively numbered opaque and sealed envelopes with treatment assignment cards were created using a 1:1 ratio with balanced blocks of randomly varying size. As patients became eligible, the envelopes were opened in consecutive order in the presence of the patient.

Description of Interventions 

Two typical usual care management strategies for patients with musculoskeletal chest pain were chosen for this study: chiropractic treatment that included spinal manipulation and self-management as an example of minimal intervention.

Participants in the chiropractic treatment group were assigned to 1 of 8 experienced chiropractors in their local community. Based on a combination of case history, clinical findings, and pragmatic, daily clinical practice, each chiropractor chose an individual treatment strategy accommodating the age and physical condition of each patient. However, treatment had to include high-velocity, low-amplitude manipulation directed toward the thoracic and/or cervical spine and could be combined with any of the following at the discretion of the treating clinician: joint mobilization, soft tissue techniques, stretching, stabilizing or strengthening exercises, heat or cold treatment, and advice. A maximum of 10 treatment sessions of approximately 20 minutes' duration each, 1 to 3 times per week for 4 weeks, was allowed. If the patient was pain-free sooner, the treating chiropractor was free to discharge the patient. The chiropractors recorded the types of treatment given after each session, as well as any adverse effects experienced by the patient.

Immediately following group allocation, the study clinician gave participants in the self-management group a 15-minute consultation consisting of reassurance and advice directed toward promoting self-management. The content of this consultation had been designed with a senior cardiologist (TH). The study clinician told participants that their chest pain generally had a benign, self-limiting course and, based on the clinical evaluation, gave individual instructions regarding posture and 2 to 3 home exercises aimed at increasing spinal movement or muscle stretch. She instructed participants to seek medical attention for reevaluation (general practitioner, cardiology, or emergency department) in the event of severe or unfamiliar chest pain. Furthermore, participants in this group were asked to refrain from seeking any type of manual treatment directed toward the muscles and joints of the thorax for the following 4 weeks.

Follow-Up Procedures 

In the chiropractic group, a receptionist in the waiting room administered a self-report questionnaire to the patients at 4 weeks post-randomization and after the final chiropractic treatment session. The questionnaire was returned immediately in a sealed envelope. At 12 weeks, a second questionnaire was administered by postal mail.

In the self-management group, patients received self-report questionnaires by postal mail at 4 and 12 weeks. Questionnaires were resent to nonresponders at 8 and 12 weeks, and at 16 and 20 weeks.

Outcome Measures 

Two primary outcome measures chosen a priori¹⁸ were (1) change in pain intensity from baseline to follow-up (“Rate your worst chest pain during the last seven days”), offering at both times an 11-point numeric rating scale, and (2) self-perceived change in chest pain at follow-up (“How is your chest pain now compared with what it was before you received treatment in this study?”) using a 7-point ordinal scale with response categories ranging from “much worse” to “much better,” with the category “unchanged” in the middle.

Secondary outcome measures included improvement in quality of life based on the Medical Outcomes Study Short Form 36-Item Health Survey (SF-36) score,¹⁹, ²⁰ 5 measures of change in pain intensity (“chest pain now,” “average chest pain,” “thoracic spine pain,” “neck pain,” and “shoulder-arm pain” reported as average intensities during the previous week), self-perceived change in general health, and self-perceived effect of treatment. Description of the outcome measures was reported elsewhere.¹⁸

Sample Size 

The sample size calculation was based on the results of the nonrandomized study by Christensen et al,¹⁵ in which self-perceived change in chest pain was assessed using an ordinal 5-point scale. A sample of 120 patients in the current study provided an 81% power for a Wilcoxon rank sum test to detect a shift in distribution of the improvement in chest pain from 0%/5%/25%/45%/25% to 1%/10%/40%/40%/9%, which corresponded to the findings by Christensen et al. However, as the current study used a 7-point scale, to ensure consistency in findings, a confirmatory analysis was performed. After obtaining the results for the 4-week assessment of self-perceived change in chest pain from 31 patients, we examined the difference in distribution between the 2 groups without knowing group membership. Taking these distributions as true, a sample of 120 patients therefore provided an 80% power for a Wilcoxon rank sum test to detect a shift in the distribution, thus confirming the sample size.

Statistical Analysis 

Baseline characteristics were reported as percentages for binary and categorical variables, and as mean values and standard deviations for continuous variables. The prevalence of pain at baseline and at 4 and 12 weeks of follow-up was calculated in absolute numbers and percentages, and compared within and between groups. Within groups, we used the McNemar test to compare follow-up with baseline; and between groups, we used the Wilcoxon rank sum test. The analysis of the secondary outcome measures “change in pain intensity” was restricted to patients with pain at baseline. We computed mean values and standard deviations in each treatment group and the differences in mean values with a t test–based 95% confidence interval. Significance was assessed using changes adjusted for baseline differences using analysis of covariance. Ratings of the self-perceived change in chest pain were reported in absolute numbers and percentages and compared between the 2 groups using a Wilcoxon rank sum test. The statistical analyses were performed on the basis of the intention-to-treat principle; that is, patients were analyzed in the group to which they were allocated. Both as-treated analysis and per-protocol analysis (where all patients with deviations from the protocol were excluded) were performed. Analyses were performed using STATA (Stata Statistical Software, release 9.2; Stata Corp, College Station, TX).

Back to Article Outline

Results 

Participants 

A detailed summary of patient recruitment, participation, and attrition is provided in Figure 1. In summary, 458 patients were eligible for baseline evaluation; and 309 consented to participation. Of these, 115 met the inclusion criteria and were randomized: 59 to the chiropractic treatment group and 56 to the self-management group. Demographic and clinical characteristics of randomized participants are summarized in Table 1.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 1. 

  Participant flow through trial.

Table 1. Baseline patient characteristics

	Chiropractic treatment (n = 59)	Self-management (n = 56)	Total (N = 115)	Pa
Age, y ± SD	51.4 ± 10.0	50.8 ± 12.1	51.1 ± 11.0	.77
Female, n (%)	26 (44.0)	22 (39.3)	48 (41.7)	.60
Body mass index, kg/m2 ± SD	27.0 ± 4.6	27.8 ± 4.7	27.4 ± 4.7	.38
Systolic blood pressure, mm Hg ± SD	135.5 ± 18.2	136.2 ± 16.9	135.8 ± 17.5	.82
Diastolic blood pressure, mm Hg ± SD	85.2 ± 10.8	87.0 ± 9.4	86.1 ± 10.1	.34
Maximum chest painb ± SD	6.7 ± 2.4	5.9 ± 2.3	6.3 ± 2.4	.076
SF-36
Physical health component scale ± SD	44.5 ± 8.8	44.2 ± 8.9	44.4 ± 8.8	.87
Mental health component scale ± SD	48.4 ± 8.9	45.8 ± 12.3	47.2 ± 10.7	.21
Risk factors of ischemic heart diseasec
Current smoker, n (%)	16 (29.6)	16 (28.6)	32 (29.1)	.90
Hypercholesterolemia, n (%)	15 (25.9)	15 (26.8)	30 (26.3)	.91
Hypertension, n (%)	12 (20.7)	15 (26.8)	27 (23.7)	.44
Diabetes, n (%)	1 (1.7)	3 (5.5)	4 (3.5)	.28
Previous episodes of chest pain, n (%)	39 (67.2)	41 (73.2)	80 (70.2)	.49
Engaging in daily light exercise, n (%)	25 (49.0)	22 (44.9)	47 (47.0)	.68
Engaging in daily heavy exercise, n (%)	3 (13.0)	2 (7.7)	5 (10.2)	.54
Married/living with someone, n (%)	51 (86.4)	50 (90.9)	101 (88.6)	.45
Working, n (%)	44 (75.9)	35 (63.6)	79 (69.9)	.16
College graduate, n (%)	16 (27.6)	16 (30.2)	32 (28.8)	.76
Expectation of treatmentd
Chiropractic treatment ± SD	2.2 ± 0.5	2.1 ± 0.6	2.1 ± 0.6	.52
Self-management ± SD	2.3 ± 0.5	2.3 ± 0.6	2.3 ± 0.5	.81

Data are expressed as mean values ± standard deviations or absolute numbers and relative frequencies (in parentheses).

Randomization resulted in 2 groups clinically comparable on baseline values. One participant in the chiropractic treatment group was dismissed without treatment by the treating chiropractor at the first visit because he found no indication for manual therapy, and one patient discontinued treatment because of time constraints. The 57 participants who completed their treatment course were seen by the chiropractors on average 7 times (range, 2-10 consultations). In the self-management group, all completed the information session.

Chiropractic Treatment 

All patients in the chiropractic treatment group received high-velocity, low-amplitude spinal manipulative therapy, most often directed toward the midthoracic region (T4 to T6 with adjoining costae) and the lower cervical spine (C6-C7). Trigger point therapy and massage were the second most commonly used treatment modalities (n = 56 patients, 98%). These were most often applied in the anterior, left side of the thorax and the trapezius muscles. In addition, a wide range of manual modalities, exercises, and advice was given and adapted to each individual patient. Adverse effects, affecting 44 patients, were transient and benign in nature, most commonly in the form of locally increased tenderness, headache, or fatigue. No serious adverse effects lasting longer than 24 hours were reported.

Prevalence of Pain 

At 4 and 12 weeks in both groups, the reduction in number with worst chest pain was statistically significant when compared with baseline, with a further decrease from 4 to 12 weeks (Table 2). At both follow-up points, there was no statistically significant difference in the proportion who had improved.

Table 2. Relative frequencies of presence of pain at baseline and follow-up with within-group comparisons and absolute numbers and relative frequencies of change in presence of pain with between-group comparisons

Both the 4- and 12-week follow-ups are considered. All patients responding at the corresponding follow-up time point are included.

With regard to the 5 secondary outcomes (Table 2), especially in the chiropractic treatment group, similar patterns were observed for chest pain now and average chest pain, that is, significant reductions in absolute numbers from baseline to 4 and 12 weeks, respectively, but without significant differences between groups. For thoracic spine pain, there was in the chiropractic treatment group no change at 4 weeks, but a significant decrease at 12 weeks, whereas in the self-management group, there were small insignificant changes at both time points. The difference between the treatment groups reached significance at 12 weeks. For neck pain, there was in both groups a statistically nonsignificant decrease at 4 weeks and only small nonsignificant changes at 12 weeks compared with baseline; and for shoulder-arm pain, there were significant decreases in both groups at 4 weeks, but at 12 weeks only in the chiropractic treatment group (Table 2).

Numeric Change in Pain Intensity 

At both 4 and 12 weeks, the largest decrease in worst chest pain was seen in the chiropractic treatment group, with adjusted differences between the groups of 0.78 and 1.10, respectively, the latter being statistically significant (P = .022) (Table 3).

Table 3. Numeric change in pain intensity from baseline to follow-up in patients with pain at baseline

All patients responding at the corresponding time point are included. CI, Confidence interval.

For the secondary outcomes, a tendency toward a larger decrease in pain intensity was seen in the chiropractic treatment group in 3 of 5 measures (chest pain now, average chest pain, and neck pain) at 4 weeks and in the self-management group in 2 of 5 measures (thoracic spine and shoulder-arm pain). At 12 weeks, a tendency to more pronounced differences between groups was seen, as the chiropractic treatment group experienced a larger decrease in 4 of 5 pain measures. None of these differences however were statistically significant.

Self-Perceived Change in Chest Pain 

At 4 weeks, we observed statistically significant better ratings in the chiropractic treatment group. Eighty-two percent (n = 44) rated their chest pain as “better” or “much better” compared with 60% (n = 28) in the self-management group. Seven percent in the chiropractic treatment group (n = 4) vs 32% (n = 15) in the self-management group rated their chest pain as unchanged. One participant in the chiropractic treatment group rated the pain as “a little worse” vs none in the self-management group. No participants reported “worse” or “much worse” (Table 4).

Table 4. Patients' self-perceived change in pain and general health, and patients' perception of treatment effect at 4 and 12 weeks of follow-up

Variable	4 wk			12 wk
	Chiropractic treatment	Self-management	Pa	Chiropractic treatment	Self-management	Pa
	n (%)	n (%)		n (%)	n (%)
How is your chest pain now compared to before treatment?
Much better	28 (51.9)	11 (23.4)	.0013	23 (45.1)	17 (35.4)	.12
Better	16 (29.6)	17 (36.2)		14 (27.5)	8 (16.7)
A little better	5 (9.3)	4 (8.5)		8 (15.7)	14 (29.2)
Unchanged	4 (7.4)	15 (31.9)		5 (9.8)	9 (18.8)
A little worse	1 (1.9)	0		1 (2.0)	0
Worse	0	0		0	0
Much worse	0	0		0	0
How is your general health now compared to before treatment?
Much better	15 (27.3)	5 (10.6)	.0004	12 (23.5)	10 (21.7)	.38
Better	23 (41.8)	10 (21.3)		18 (35.3)	11 (23.9)
A little better	7 (12.7)	11 (23.4)		8 (15.7)	10 (21.7)
Unchanged	9 (16.4)	20 (42.6)		12 (23.5)	15 (32.6)
A little worse	1 (1.8)	1 (2.1)		1 (2.0)	0
Worse	0	0		0	0
Much worse	0	0		0	0
Did the treatment help?
Helped a lot	35 (63.6)	14 (30.0)	.0009	27 (52.9)	18 (37.5)	.17
Helped some	17 (30.9)	27 (57.5)		16 (31.4)	21 (43.8)
Did not help	2 (3.6)	6 (12.8)		7 (13.7)	7 (14.6)
Got worse	1 (1.8)	0		1 (2.0)	2 (4.2)

Absolute and relative numbers are given.

At 12 weeks, the difference in self-perceived change in chest pain was no longer statistically significant because the self-management group rated slightly more positive than at 4 weeks, with 35% (n = 17) rating “much better” and 17% (n = 8) rating “better” (vs 23% [n = 11] and 36% [n = 17] at 4 weeks). Ten percent (n = 5) in the chiropractic treatment and 19% (n = 9) in the self-management group were unchanged. Self-perceived change in general health and self-perceived effect of treatment showed similar results, with statistically nonsignificant differences in favor of the chiropractic treatment group (Table 4).

Quality of Life Scores (SF-36) 

Baseline scores are presented in Table 1. Compared with an age-matched background population, patients in this study scored between 1.27 (bodily pain) and 0.28 SD (social function) lower at baseline. Except for social function in the chiropractic treatment group at 4 weeks, both groups experienced improvements in all 8 domains at 4 and 12 weeks and scored between 0.08 SD lower (chiropractic treatment group at 12 weeks, physical function) and 0.76 SD lower (chiropractic treatment group at 4 weeks, role physical) than the age-matched background population. Between-group differences neither reached significance nor showed a common trend (data not shown).

Analysis of Sensitivity of Results 

Despite requesting the participants in the self-management group to refrain from manual therapy, 5 participants received chiropractic treatment within 4 weeks of baseline. In addition, 1 patient in the chiropractic treatment group did not initiate treatment. However, the results of the as-treated and per-protocol analyses (data not shown) were in line with the results from the intention-to-treat analyses and did not change the final conclusions.

Back to Article Outline

Discussion 

To our knowledge, this is the first report on a randomized controlled trial evaluating the effect of chiropractic treatment of musculoskeletal chest pain in patients dismissed from a cardiac emergency department without a diagnosis of ACS when this is compared with a minimal intervention. In both groups, we found a decrease in pain intensity and in the proportion of patients with pain, self-perceived positive changes in chest pain, and increases in SF-36, all reflecting better health and well-being. At statistically significant levels, chiropractic treatment was favored by the primary outcomes “self-perceived change in chest pain” at 4 weeks and “numeric change in pain intensity” at 12 weeks. The majority of the nonsignificant results was also in favor of the chiropractic treatment group.

Interpretation of Results 

Both groups experienced reductions in pain intensity and positive self-perceived treatment effects, which could be the result of regression toward the mean or the course of natural history. The natural history and degree of symptom turnover of musculoskeletal chest pain are poorly understood. Studies addressing the long-term course of patients with musculoskeletal chest pain have not been identified, but studies of patients with noncardiac chest pain have addressed the issue. Although considered a benign condition, it seems that noncardiac chest pain has the ability to critically intrude into daily life. Recurrent episodes of pain are estimated to affect 50% to 90% of patients with noncardiac chest pain.⁶, ²¹, ²², ²³, ²⁴ Continued episodes of pain are associated with disability²⁵, ²⁶; work absenteeism²¹, ²², ²⁴, ²⁵; reduced quality of life in terms of mental health, vitality, depression, and anxiety⁶, ²²; as well as increased number of contacts with the overall health care system.²³, ²⁷ This is comparable with the baseline findings in this study with patients showing poorer health as measured with SF-36 compared with an age-matched background population.

The tendency of results in favor of the chiropractic treatment group was consistent at both follow-up points, and a closer look reveals interesting patterns. Both groups experienced reduction in worst chest pain intensity (Table 3), but with a tendency to increasing group differences over time, driven by increasing reductions in the chiropractic treatment group. This general tendency was actually noted for 4 of 6 measures of pain intensity. A slight variation could be observed for average thoracic spine pain, with a smaller reduction in the chiropractic group at 4 weeks and a reversed pattern at 12 weeks. This probably reflects less local soreness, which was often reported in the chiropractic treatment group. For the subjective outcomes, the general tendency was opposite: the group differences favoring chiropractic treatment decrease over time. This was driven by a more positive rating in the self-management group at 12 compared with 4 weeks. As the level of pain intensity generally remained constant in the self-management group from 4 to 12 weeks, the change in self-perceived change in pain may reflect that these patients gradually adapt to their situation despite continued pain. These results indicate that the time point at which the participants are asked about symptoms and treatment effect is important. Pain perception appears to be a dynamic process, and pain intensity and self-perceived change in chest pain may be rated independently of each other. Although group differences were small, the improvement in pain intensity in both groups is substantial. One may argue that chiropractic treatment is an “add on,” perhaps leading to faster recovery than without treatment.

Strengths 

We chose to undertake a randomized controlled trial to identify the effect of chiropractic treatment compared with self-management, as these 2 management strategies resembled usual conditions of care. This pragmatic approach enabled us to identify any effect resulting from the overall chiropractic care.

The chiropractic treatment was given in community-based chiropractic clinics by 8 different chiropractors each treating their own group of participants. Therefore, observed effects were independent of a single care provider. Moreover, personalized treatment given at the discretion of each chiropractor reflects standard practice; and hence, for both reasons, we can expect that our results can be generalized with high validity.

As recommended for studies of pain,²⁸ we assessed several patient-rated variables. Pain intensity and self-perceived change in chest pain were chosen a priori as the primary outcomes because these are considered the most important outcomes in patients with a variety of pain syndromes, along with quality of life.²⁸ Global rating scales are regarded as clinically relevant and responsive to measure patients' perceived recovery.²⁹

The diagnosis of musculoskeletal chest pain was based on systematic patient assessment using a standardized protocol shown to have substantial interobserver agreement.¹⁷ The issue of validity of this diagnosis has been addressed previously by us in patients with stable angina pectoris in whom an experienced clinician could fairly convincingly identify a subset of patients with musculoskeletal chest pain.¹⁵ In the current study, positive response to treatment targeted at structures believed to be pain generators (muscles and joints of the cervicothoracic spine and chest wall) gave further support to the validity of the diagnosis. In addition, we feel that our multidisciplinary research team comprising experts in cardiology, clinical physiology, chiropractic, and biostatistics may have served as a shield against one-eyed interpretations.

Limitations 

Musculoskeletal chest pain is a clinical diagnosis without a criterion standard, and its presence is difficult to confirm. Thus, the diagnosis is susceptible to interobserver variation; and this limits the generalizability of our study results, as they probably depend on the correct identification of patients who can benefit from chiropractic treatment.

The decrease in chest pain in the self-management group indicates that chest pain will disappear in some patients without any treatment; and therefore, the results in both groups are probably affected by regression toward the mean. However, this cannot explain group differences.

The pragmatic design of the current study did not allow for standardization of treatment or identification of active ingredients of the intervention, and we cannot exclude that the treatment effect in favor of the chiropractic treatment groups was not caused by provider attention. This question has to be addressed by further research.

Generally, participants had positive expectations within both treatment groups as opposed to patients who declined participation, who often expressed negative expectations toward chiropractic treatment. As a consequence, study results may not be representative of nonparticipants. The same tendency was noted previously by Christensen et al¹⁶ who found that refusal to participate was due to either a negative attitude toward chiropractic treatment or an ongoing belief that heart disease had not been ruled out even when the results of all cardiac examinations were normal.

The protocol used to identify patients with musculoskeletal chest pain was not designed to classify or exclude other potential causes of chest pain; and thus, we could not allege musculoskeletal chest pain as the sole cause of chest discomfort. Gastroesophageal reflux disease is probably the most common differential diagnosis¹¹; but because of constraints, specific examination for gastroesophageal reflux disease was not possible, nor did we examine for overlap with other causes of nonspecific chest pain, for example, anxiety and depression.⁵, ³⁰ The presence of these causes may have limited the benefit of the interventions in some patients.

We included patients with an acute episode of chest pain and found that the population comprised patients with both first time and repeated episodes. The 2 groups may not respond equally well to spinal manipulative therapy, and focusing on patients with first-time symptoms may reveal larger treatment effects.³¹

Only 115 patients were included in the study, and approximately 15% of participants in the self-management group at 4 weeks and in both groups at 12 weeks did not return their questionnaires. The inability to meet the required sample size and loss to follow-up may have resulted in underpowering of the study and the risk of a type II error.

Adverse effects were only queried in the chiropractic treatment group; and finally, there is a general uncertainty regarding the outcome measures of the disease-unspecific measure “pain” and its clinical implications.

Lessons Learned and Questions to Be Answered 

Being a new area of research, many assumptions behind the rationale for the study were based on the experience of research on nonspecific neck and low back pain. Musculoskeletal chest pain may be (partly) caused by neck and thoracic spine dysfunction, and recent evidence indicates that thoracic spine pain may have a similar etiology and natural course as that of neck and low back pain.³² In neck and low back pain caused by joint dysfunction, spinal manipulative therapy is assumed to improve joint dysfunction³³ and is considered a beneficial treatment approach.³⁴, ³⁵

The combined research effort of participants from such diverse fields as chiropractic and cardiology was a challenge and an eye-opener to all involved. These results illustrate how open-mindedness together with a multidisciplinary approach can help bring forward clinically relevant answers to complex areas such as undifferentiated chest pain.

Discomfort, shortness of breath, anxiety, fear, and depression are other aspects of an acute chest pain episode that may lead to or contribute to admission in this category of patients. Such aspects were not explored in this study. Clarification of what aspects of the pain episode patients consider important may be needed to schedule appropriate treatment.

Our study suggests that chiropractic treatment might at least lead to a faster recovery in patients with acute musculoskeletal chest pain. To minimize suffering and overall cost in these patients, the scale of the problem in terms of number of patients concerned, the documented consequences in daily life and repeated contact with the health care system for those affected, and the limited cost of a short series of chiropractic treatments appear to be good reasons to investigate whether offering chiropractic treatment is justified. In addition, the handling of patients with noncardiac chest pain after discharge from the cardiologic unit is challenging for general practitioners; and our results suggest that a trial of chiropractic treatments may be a feasible option in the absence of other obvious causes for chest pain.

Back to Article Outline

Conclusions 

We conducted what we believe is the first randomized controlled trial assessing chiropractic treatment of acute musculoskeletal chest pain compared with self-management in patients dismissed from an emergency clinic without a diagnosis of ACS. With both regimens, we observed substantial decreases in chest pain; however, there were statistically significant and nonsignificant between-group differences in favor of the chiropractic group. The active ingredients in either treatment are unknown, and regression toward the mean probably plays a role. The tendencies and trends in favor of chiropractic treatment should encourage further research to examine exactly what role chiropractic treatment can play in patients with acute chest pain of musculoskeletal origin.

Back to Article Outline

Funding Sources and Potential Conflicts of Interest 

No conflicts of interest were reported for this study. This project was supported by the Foundation of Chiropractic Research and Postgraduate Education, Denmark, and Funen County, Denmark.

Back to Article Outline

References 

1. Nawar EW, Niska RW, Xu J. National Hospital Ambulatory Medical Care Survey: 2005 emergency department summary. Adv Data. 2007;1–32
   • View In Article
2. Goodacre S, Cross E, Arnold J, Angelini K, Capewell S, Nicholl J. The health care burden of acute chest pain. Heart. 2005;91:229–230
   • View In Article
3. Murphy NF, MacIntyre K, Capewell S, Stewart S, Pell J, Chalmers J, et al. Hospital discharge rates for suspected acute coronary syndromes between 1990 and 2000: population based analysis. BMJ. 2004;328:1413–1414
   • View In Article
   • CrossRef
4. Eslick GD, Coulshed DS, Talley NJ. Review article: the burden of illness of non-cardiac chest pain. Aliment Pharmacol Ther. 2002;16:1217–1223
   • View In Article
   • MEDLINE
   • CrossRef
5. How J, Volz G, Doe S, Heycock C, Hamilton J, Kelly C. The causes of musculoskeletal chest pain in patients admitted to hospital with suspected myocardial infarction. Eur J Intern Med. 2005;16:432–436
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (87 KB)
   • CrossRef
6. Eslick GD, Talley NJ. Natural history and predictors of outcome for non-cardiac chest pain: a prospective 4-year cohort study. Neurogastroenterol Motil. 2008;20:989–997
   • View In Article
   • CrossRef
7. Goodacre S, Mason S, Arnold J, Angelini K. Psychologic morbidity and health-related quality of life of patients assessed in a chest pain observation unit. Ann Emerg Med. 2001;38:369–376
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (81 KB)
   • CrossRef
8. Fagring AJ, Gaston-Johansson F, Danielson E. Description of unexplained chest pain and its influence on daily life in men and women. Eur J Cardiovasc Nurs. 2005;4:337–344
   • View In Article
   • MEDLINE
   • CrossRef
9. Fagring AJ, Kjellgren KI, Rosengren A, Lissner L, Manhem K, Welin C. Depression, anxiety, stress, social interaction and health-related quality of life in men and women with unexplained chest pain. BMC Public Health. 2008;8165
   • View In Article
10. Kahn SE. The challenge of evaluating the patient with chest pain. Arch Pathol Lab Med. 2000;124:1418–1419
    • View In Article
    • MEDLINE
11. Eslick GD, Coulshed DS, Talley NJ. Diagnosis and treatment of noncardiac chest pain. Nat Clin Pract Gastroenterol Hepatol. 2005;2:463–472
    • View In Article
    • MEDLINE
    • CrossRef
12. Stochkendahl MJ, Christensen HW. Chest pain in focal musculoskeletal disorders. Med Clin North Am. 2010;94:259–273
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (499 KB)
    • CrossRef
13. Jensen S. Musculoskeletal causes of chest pain. Aust Fam Physician. 2001;30:834–839
    • View In Article
    • MEDLINE
14. Polkinghorn BS, Colloca CJ. Chiropractic management of chronic chest pain using mechanical force, manually assisted short-lever adjusting procedures. J Manipulative Physiol Ther. 2003;26:108–115
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (143 KB)
    • CrossRef
15. Christensen HW, Vach W, Gichangi A, Manniche C, Haghfelt T, Høilund-Carlsen PF. Cervicothoracic angina identified by case history and palpation findings in patients with stable angina pectoris. J Manipulative Physiol Ther. 2005;28:303–311
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (172 KB)
    • CrossRef
16. Christensen HW, Vach W, Gichangi A, Manniche C, Haghfelt T, Høilund-Carlsen PF. Manual therapy for patients with stable angina pectoris: a nonrandomized open prospective trial. J Manipulative Physiol Ther. 2005;28:654–661
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (125 KB)
17. Brunse MH, Stochkendahl MJ, Vach W, Kongsted A, Poulsen E, Hartvigsen J, et al. Examination of musculoskeletal chest pain—an inter-observer reliability study. Man Ther. 2010;15:167–172
    • View In Article
    • CrossRef
18. Stochkendahl MJ, Christensen HW, Vach W, Høilund-Carlsen PF, Haghfelt T, Hartvigsen J. Diagnosis and treatment of musculoskeletal chest pain: design of a multi-purpose trial. BMC Musculoskelet Disord. 2008;940
    • View In Article
19. Bjorner JB, Kreiner S, Ware JE, Damsgaard MT, Bech P. Differential item functioning in the Danish translation of the SF-36. J Clin Epidemiol. 1998;51:1189–1202
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (222 KB)
    • CrossRef
20. Bjorner JB, Damsgaard MT, Watt T, Groenvold M. Tests of data quality, scaling assumptions, and reliability of the Danish SF-36. J Clin Epidemiol. 1998;51:1001–1011
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (189 KB)
    • CrossRef
21. Spalding L, Reay E, Kelly C. Cause and outcome of atypical chest pain in patients admitted to hospital. J R Soc Med. 2003;96:122–125
    • View In Article
    • MEDLINE
    • CrossRef
22. Chambers J, Bass C. Chest pain with normal coronary anatomy: a review of natural history and possible etiologic factors. Prog Cardiovasc Dis. 1990;33:161–184
    • View In Article
    • MEDLINE
    • CrossRef
23. Panju A, Farkouh ME, Sackett DL, et al. Outcome of patients discharged from a coronary care unit with a diagnosis of “chest pain not yet diagnosed.”. Can Med Assoc J. 1996;155:541–546
    • View In Article
24. Juelsgaard P, Ronnow Sand NP. Somatic and social prognosis of patients with angina pectoris and normal coronary arteriography: a follow-up study. Int J Cardiol. 1993;39:49–57
    • View In Article
    • MEDLINE
    • CrossRef
25. Achem SR, DeVault KR. Recent developments in chest pain of undetermined origin. Curr Gastroenterol Rep. 2000;2:201–209
    • View In Article
    • MEDLINE
    • CrossRef
26. Jerlock M, Welin C, Rosengren A, Gaston-Johansson F. Pain characteristics in patients with unexplained chest pain and patients with ischemic heart disease. Eur J Cardiovasc Nurs. 2006;6:130–136
    • View In Article
    • MEDLINE
    • CrossRef
27. Tew R, Guthrie EA, Creed FH, Cotter L, Kisely S, Tomenson B. A long-term follow-up study of patients with ischaemic heart disease versus patients with nonspecific chest pain. J Psychosom Res. 1995;39:977–985
    • View In Article
    • Abstract
    • Full-Text PDF (570 KB)
    • CrossRef
28. Turk DC, Dworkin RH, Allen RR, Bellamy N, Brandenburg N, Carr DB, et al. Core outcome domains for chronic pain clinical trials: IMMPACT recommendations. Pain. 2003;106:337–345
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (119 KB)
    • CrossRef
29. Lauridsen HH, Hartvigsen J, Korsholm L, Grunnet-Nilsson N, Manniche C. Choice of external criteria in back pain research: does it matter? Recommendations based on analysis of responsiveness. Pain. 2007;131:112–120
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (205 KB)
    • CrossRef
30. Eslick GD, Jones MP, Talley NJ. Non-cardiac chest pain: prevalence, risk factors, impact and consulting—a population-based study. Aliment Pharmacol Ther. 2003;17:1115–1124
    • View In Article
    • MEDLINE
    • CrossRef
31. Bronfort G, Haas M, Evans RL, Bouter LM. Efficacy of spinal manipulation and mobilization for low back pain and neck pain: a systematic review and best evidence synthesis. Spine J. 2004;4:335–356
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (302 KB)
    • CrossRef
32. Leboeuf-Yde C, Nielsen J, Kyvik KO, Fejer R, Hartvigsen J. Pain in the lumbar, thoracic or cervical regions: do age and gender matter? A population-based study of 34,902 Danish twins 20-71 years of age. BMC Musculoskelet Disord. 2009;10:39
    • View In Article
    • CrossRef
33. Bronfort G, Evans R, Nelson B, Aker PD, Goldsmith CH, Vernon H. A randomized clinical trial of exercise and spinal manipulation for patients with chronic neck pain. Spine (Phila Pa 1976 ). 2001;26:788–797
    • View In Article
    • MEDLINE
    • CrossRef
34. Gross AR, Goldsmith C, Hoving JL, et al. Conservative management of mechanical neck disorders: a systematic review. J Rheumatol. 2007;34:1083–1102
    • View In Article
35. Underwood M. Low back pain—early management of persistent non-specific low back pain. NICE clinical guideline 88. Developed by the National Collaborating Centre for Primary Care [Internet]. March 2010 [cited 2010 Oct].Available from: www.nice.org.uk/CG88.
    • View In Article