Comparison of ibuprofen, cyclobenzaprine or both in patients with acute cervical strain: a randomized controlled trial

EM Advances

S. Mustafah Khwaja, DO; Max Minnerop, MD; Adam J. Singer, MD

From the Department of Emergency Medicine, Stony Brook University, Stony Brook, NY

CJEM 2010;12(1):39-44


This article has been peer reviewed.

Objective: We compared pain severity and time to resumption of activities in patients with cervical strains treated with a nonsteroidal anti-inflammatory drug (NSAID), a centrally acting muscle relaxant or both. Methods: We performed a double blinded, randomized controlled trial of adults with cervical strains from motor vehicle collisions or from falls who presented to a suburban academic emergency department (ED). Patients were randomly assigned to receive ibuprofen 800 mg, cyclobenzaprine 5 mg or both, 3 times daily as needed for up to 7 days. Outcome measures included a pain score on a 100 mm visual analog scale, pain relief scores, the time to resumption of normal activities, the use of rescue medications, and adverse outcomes. We used repeated measures analysis of variance to compare pain relief over time. Our sample size of 20 patients in each group had a power of 80% to detect a difference of 15 mm in pain relief scores between the highest and lowest groups. Results: We randomly assigned 61 patients to receive ibuprofen (n = 20), cyclobenzaprine (n = 21) or both (n = 20). Mean (standard deviation) age was 34 (11) years; 58% were women and 72% were white. Although pain scores improved over time in all groups, there were no significant differences between the groups in any of the outcome measures. The rate of adverse events was also similar between groups. Conclusion: Our study suggests that there is little benefit to routinely using or adding cyclobenzaprine to NSAIDs for ED patients with acute cervical strain.


Objectif : Nous avons comparé la gravité de la douleur et le temps de reprise des activités chez les patients présentant une entorse du rachis cervical et ayant été traités avec un anti-inflammatoire nonstéroïdien (AINS) ou un relaxant musculaire à action centrale, ou les deux. Méthodes : Nous avons réalisé un essai contrôlé randomisé à double insu auprès d’adultes qui ont consulté le service d’urgence d’un centre hospitalier universitaire de banlieue pour une entorse du rachis cervical liée à un accident de véhicule automobile ou à une chute. Les participants ont été répartis de façon aléatoire pour recevoir 800 mg d’ibuprofène, 5 mg de cyclobenzaprine, ou les deux, 3 fois par jour, au besoin, pour un maximum de 7 jours. Les principales mesures des résultats comprenaient le score de douleur sur une échelle visuelle analogique de 100 mm, les scores de soulagement de la douleur, le délai de reprise des activités nor males, l’utilisation de médicaments de secours et les événements indésirables. Nous avons utilisé une analyse de variance à mesures répétées pour comparer le soulagement de la douleur dans le temps. Chaque groupe de 20 patients était capable à 80 % de détecter une différence de 15 mm quant au score de gravité de la douleur entre le groupe le plus élevé et le groupe le plus bas. Résultats : Soixante et un patients ont été répartis de façon aléa toire pour recevoir de l’ibuprofène (n = 20), de la cyclobenzaprine (n = 21) ou les deux (n = 20). L’âge moyen était de 34 ans (écart type, 11); 58 % étaient des femmes et 72 % étaient Blancs. Bien que les scores de douleur se soient améliorés au fil du temps pour tous les groupes, il n’y avait pas de différences significatives entre les groupes pour aucune des mesures de résultats. Le taux d’événements indésirables était également similaire pour tous les groupes. Conclusion : Notre étude indique qu’il y a peu d’avantages à prescrire la cyclobenzaprine ou à l’ajouter de façon routinière aux AINS chez les patients se présentant à l’urgence pour une entorse du rachis cervical.


In each year there are 13 million motor vehicle collisions (MVCs) in the United States and an estimated 500 000 to 1 million result in cervical strains.1 The diagnosis, treatment and insurance payments directly related to cervical strains cost an estimated US$4.5 billion to $29 billion annually.2 The injury is caused by sudden acceleration or deceleration of the head relative to the trunk.3 The cause of pain associated with cervical strains is poorly understood. Studies using human cadavers suggest that the occipital pain is caused by the stretch of dorsal root ganglia of C1 and C2, and that the characteristic lower cervical pain is caused by compression of the cervical facet joints.4 Symptoms of cervical strain include pain in the neck, shoulder, arm, head or jaw, dizziness, tinnitus, and difficulties with memory or concentration.5

Pain and muscle spasm are closely related in musculoskeletal pain syndromes. The initial injury is believed to cause reflex spasm of the affected muscles. This muscle spasm results in further pain propagating a vicious cycle of increased pain and discomfort.6 As a result, many physicians prescribe muscle relaxants in addition to analgesics and anti-inflammatory agents.7,8 Previous studies of muscle relaxants in combination with nonsteroidal anti-inflammatory drugs (NSAIDs) for lower back pain have demonstrated minor advantages of the combination therapy over single agent therapy.9-12 However, most of these studies recruited nonemergency patients, were industry funded and took place more than 20 years ago, when bed rest was recommended rather than early mobility. Other treatment modalities such as physiotherapy and exercise programs have not been studied in the emergency department (ED) setting.

The objective of the current study was to determine whether a centrally acting muscle relaxant (cyclobenzaprine) reduces pain and speeds recovery when compared with NSAIDs alone or in combination for ED patients with acute cervical strains.


Study design

We used a prospective, double blinded, randomized clinical trial (no. NCT00790270) to test the study hypothesis. All patients provided written informed consent. Our institutional review board approved the study.


We conducted the study in the ED of a suburban teaching hospital with an accredited emergency medicine residency program and an annual census of 75 000. The study site is a regional level 1 trauma centre.


A convenience sample of patients was enrolled Monday through Saturday from 8:00 am to 10:00 pm when a trained research assistant was on site.13 Patients were eligible for enrolment if they presented to the ED within 24 hours of an MVC or fall, were 18 years of age or older, and answered the question "Do you have any neck pain?" in the affirmative. The question was initially asked by the triage nurse and treating physician, and repeated by a trained research assistant immediately before enrolment. The location, quality or severity of the injury did not otherwise affect eligibility for enrolment. Plain radiographs of the cervical spine were obtained in selected patients based on the criteria of the National Emergency X Radiography Utilization Study.14 We excluded patients for the following reasons: they were not available for telephone follow up; they were pregnant; they had an allergy or intolerance to any of the study drugs; they had used a monoamine oxidase inhibitor within 14 days; they had a history of arrhythmias, heart failure, recent myocardial infarction, hyperthyroidism, cervical spine fractures or dislocations, or focal neurologic findings (either central or peripheral); they were admitted to hospital; they experienced impaired cognitive function that would preclude informed consent and pain assessment. We also excluded patients with a history of gastrointestinal bleeding or substance abuse.


After giving informed consent, all participants were administered 800 mg of ibuprofen by mouth and were then randomly assigned in equal proportion to 1 of 3 treatment groups. This initial dose of ibuprofen was given to ensure that even patients randomly assigned to cyclobenzaprine alone received at least 1 dose of analgesic. We determined assignment of therapy by opening the next consecutively numbered opaque envelope. The study assignments were based on a computerized random numbers table and were prepared by personnel who were not involved in the study. Both physicians and patients were blinded to treatment allocation. Patients assigned to the first study group were treated with ibuprofen 800 mg and an inactive placebo tablet, 3 times daily by mouth. Patients assigned to the second study group were treated with a similarly appearing inactive placebo tablet and cyclobenzaprine 5 mg, 3 times daily. Patients assigned to the third group were treated with both ibuprofen 800 mg and cyclobenzaprine 5 mg, 3 times daily. All treatments were to be taken as needed for up to 7 days or until pain relief was considered adequate by the patient. There were no dose adjustments based on weights of participants.

Measures and outcomes

Research assistants recorded patient demographics, medical history, medications prescribed and initial pain scores onto standardized forms. We obtained pain scores from patients using a previously validated unhatched 100 mm visual analog scale (VAS) marked "no pain" and "most pain" at the low and high ends, respectively.15 Patients were asked to record in a daily log neck pain severity and return to work, school or other normal activities after discharge. The pain severity was to be recorded using a 100 mm VAS 30-60 minutes after taking the morning dose of the assigned treatment. A previous study16 demonstrated a high correlation between pain scores recorded at home and in the ED. Treatment was continued for 7 days or until pain relief was considered satisfactory by the patients. Patients also recorded the use of ice packs and rescue with acetaminophen 1 g every 6 hours. Use of ice packs or other nonpharmacological therapies were not considered rescue therapies. Patients were also asked to record the occurrence of nausea, vomiting and dizziness, as well as any other adverse events. The study protocol did not specify how physicians should instruct patients with regard to the use of physiotherapy or exercise programs. After 7 days, patients were to return the diaries by mail using self addressed envelopes. We conducted telephone reminders when the diaries were not received within 2 weeks.

The main outcome measure was the change in pain scores over time as measured by the pain intensity differences (PIDs), which were the differences between the baseline pain VAS and the VAS at each follow up measurement point in time; positive values of the PIDs indicate decreases in pain. Secondary outcomes were the time weighted summed pain intensity difference (SPID) score, the number of days until resumption of normal daily activities, the use of rescue medications, and adverse events. We calculated the SPID score by weighting the PIDs by the number of days since the previous observation and summing these weighted differences.

Data analysis

Continuous data are presented as means and confidence intervals (CIs) and compared with analysis of variance (ANOVA) and repeated measures ANOVA as appropriate; data were tested for normality to justify the necessary assumptions for using these tests. We present binomial data as percent frequency of occurrence with 95% CIs and we compared the data with χ2 or Fisher exact tests. We compared the primary outcome among groups using repeated measures ANOVA. A sample size of 20 patients in each group had greater than 80% power to detect a difference of 15 mm in PID scores between the highest and lowest groups assuming a common standard deviation of 15, further assuming that the mean PID of the ibuprofen alone and cyclobenzaprine alone groups were more similar than that of the combined group. Data analysis was on an intention to treat basis using SPSS 15.0 for Windows (SPSS Inc.).


During the study period 61 patients were enrolled. Their mean age (standard deviation [SD]) was 34 (11) years; 58% were women and 72% were white. Most (87%) cervical strains were the result of an MVC. The mean initial pain using VAS pain scoring was 52.8 mm (95% CI 48.2-57.5 mm). No patients were lost to follow up.

The patients were randomly assigned to receive ibuprofen (n = 20), cyclobenzaprine (n = 21) or both (n = 20). The 3 study groups were similar in baseline characteristics such as age, sex, race, initial pain scores and mechanism of injury (Table 1).

Table 1. Baseline characteristics of 61 patients included in the study, by treatment group
Characteristic No. (%) of patients* p value
Ibuprofen, n = 20 Cyclobenzaprine, n = 21 Ibuprofen + cyclobenzaprine, n = 20
Mean (SD) age, yr 32 (11) 37 (13) 34 (10) 0.33
Female 11/18 (61)‡ 14/21 (67) 9/20 (45) 0.35
White 14/20 (70) 16/21 (76) 14/20 (70) 0.84
Mean (SD) initial pain score, mm† 52 (18) 50 (8) 57 (19) 0.41
MVC 15/19 (79)§ 20/21 (95) 18/19 (95)§ 0.12
MVC = motor vehicle collision; SD = standard deviation. *Unless otherwise indicated. †Measured on a 100-mm visual analog pain scale. ‡2 patients with unspecified sex. §1 patient with unspecified cause of injury.

The changes in pain scores over time were similar among the 3 treatment groups (Fig. 1). There was a slight increase in pain in all groups the day after base line (mean PID on day 1 -4.5, -8.5 and -5.1 for the ibuprofen, cyclobenzaprine and combination groups, respectively), and the most pain relief was measured on day 7 (mean PID on day 7 was 26.0, 35.0 and 43.3). In the ibuprofen group there was a tendency toward a slower rate of pain relief over time as indicated by the repeated measures ANOVA interaction term (p = 0.06). Individually, only the PID on day 6 was significantly different among the 3 groups (p = 0.05, unadjusted for multiple testing). In all 3 study groups there was a significant reduction in pain scores over time (p < 0.001). The SPID was not significantly different across groups (Table 2). Patients who received ibuprofen, with or without cyclobenzaprine, tended to require fewer res cue medications, but this difference was not statistically significant (p = 0.17). Similarly, patients assigned to receive ibuprofen, with or without cyclobenzaprine, tended to be more likely to resume normal activities the following day, but this difference was also not significant (p = 0.11).

Figure 1

Fig. 1. Mean change in pain (pain intensity difference) over time (d). A positive value represents a decrease in pain score, by self-report on a 100-mm visual analog scale 30 to 60 minutes after morning dose of study medication. Bars represent standard errors.

Table 2. Secondary outcomes of 61 patients included in the study, by treatment group
Outcome No. (%) of patients* p value
Ibuprofen, n = 20 Cyclobenzaprine, n = 21 Ibuprofen + cyclobenzaprine, n = 20
Mean (95% CI) SPID score 65 (13-143) 118 (61-175) 131 (75-188) 0.29
Used rescue medications 9 (45) 13 (62) 9 (45) 0.17
Resumed activities next day 14 (70) 8 (38) 13 (65) 0.11
Adverse events 1 (5) 1 (5) 3 (15) 0.40
Dizziness 0   1   3  
Nausea 1   0   0  
CI = confidence interval; SPID = summed pain intensity difference. *Unless otherwise indicated.

Four patients receiving cyclobenzaprine alone or in combination reported dizziness, and 1 patient treated with ibuprofen alone reported nausea.


The results of our study demonstrate that the addition of cyclobenzaprine to ibuprofen does not result in better pain relief or earlier resumption of normal daily activities than ibuprofen alone. Moreover, the number of adverse events in the combined therapy group was slightly greater than either study group alone. Because the addition of cyclobenzaprine increases the costs and has the potential to also increase the risks of adverse events, our study suggests that there is little, if any, benefit to routinely adding cyclobenzaprine to NSAIDs in the management of ED patients with acute cervical strain.

Cyclobenzaprine is a tricyclicamine similar in structure and pharmacology to amitriptyline, and a weak inhibitor of presynaptic norepinephrine and serotonin reuptake. It has strong anticholinergic, antihistaminic and sedative properties but weak antidepressant effects. Based on animal studies, cyclobenzaprine is believed to relieve muscle spasm by acting centrally in the brain stem to reduce descending tonic discharges on γ and α motoneurons. Cyclobenzaprine may also have direct activity on spinal motoneurons.9

Other studies have also evaluated centrally acting muscle relaxants, such as cyclobenzaprine, with or with out NSAIDs, for musculotendinous injuries. A previous meta analysis performed by Browning and colleagues9 demonstrated that cyclobenzaprine was significantly better than placebo for the treatment of lower back pain. This meta analysis included 14 clinical studies, most of which were poorly randomized, unblinded and industry sponsored. Additionally, the beneficial effect of cyclobenzaprine was modest and limited to the first few days after injury. A double blind trial of diflunisal, with or without cyclobenzaprine,10 demonstrated a significant advantage of the combination therapy limited to the fourth day of treatment only. A nonblinded comparative study of naproxen, with or without cyclobenzaprine using nonvalidated outcomes,11 reported a nonsignificant improvement in function and pain in the combined therapy group. The minimal effects of muscle relaxants on pain may be explained by the suggestion that muscle spasm does not increase pain as previously believed.17

A study of 77 ED patients with diverse acute myofascial strains by Turturro and coworkers12 did not find any benefit of adding cyclobenzaprine to ibuprofen. The cyclobenzaprine did not reduce pain, and it was associated with an increased rate of central nervous system side effects (42% v. 18%). However, in that study there was no group treated with cyclobenzaprine alone and the dose of cyclobenzaprine used was 10 mg. For our study we chose to use a 5 mg dose of cyclobenzaprine, since updated recommendations by the manufacturer suggest that this lower dose is as effective yet has fewer adverse events. Another difference is that our study population was more homogenous including only patients with cervical strains.

Our findings are consistent with the clinical practice guideline on acute lower back pain problems in adults published by the Agency for Health Care Policy and Research, which concluded that skeletal muscle relax ants have not been shown to be more effective than NSAIDS, and that no additional benefit was gained by using muscle relaxants in combination with NSAIDs over use of NSAIDs alone.18


Our study has several notable limitations. This was a convenience sample of patients who were enrolled when research assistants were present, and we cannot exclude a selection bias. We also did not collect data on those patients who were not enrolled in the study for comparison. The small sample size limited our ability to detect smaller differences in outcomes (a type II error). We also did not measure the presence and the severity of any associated muscle spasm. However, this is likely to be subjective and subject to interrater variability and we could not identify validated measures of muscular spasm. Potential confounders may have been overlooked. For example, we did not measure the use of physiotherapy and exercising during the study period. We also did not assess function other than resumption of daily activities. Finally, we chose a fixed dose design, and cannot comment on the safety or effectiveness of other doses.


The addition of cyclobenzaprine to ibuprofen in the treatment of ED patients with acute cervical strains resulting from MVCs or falls does not appear to result in more effective pain relief or faster resumption of normal daily activities. Thus we do not recommend routinely adding cyclobenzaprine to ibuprofen in patients presenting to the ED with cervical strains.


  1. Dreyer SJ, Boden SD. Nonoperative treatment of neck and arm pain. Spine 1998;23:2746-54.

  2. Freeman MD, Croft AC, Rossignol AM, et al. A review and methodologic critique of the literature refuting the whiplash syndrome. Spine 1999;24:86-98.

  3. Sturzenegger M, DiStefano G, Radanov BP, et al. Presenting symptoms and signs after whiplash injury: the influence of accident mechanisms. Neurology 1994;44:688-93.

  4. Cusick JF, Pintar FA, Yoganadan N. Whiplash syndrome: kinematic factors influencing pain patterns. Spine 2001;26:1252-8.

  5. Spitzer WO, Skovron ML, Salmi LR, et al. Scientific monograph of the Quebec Task Force on Whiplash Associated Disorders: redefining "whiplash" and its management. [Erratum, Spine 1995; 20:2372.]. Spine 1995;20(Suppl):1S-73S.

  6. Roland M. A critical review of the evidence for a pain spasm pain cycle in spinal disorders. Clin Biomech (Bristol, Avon) 1986;1:102-9.

  7. Chang D, Bosco JA. Cervical spine injuries in the athlete. Bul NYU Hosp Jt Dis 2006;64:119-29.

  8. Zmurko MG, Tannoury TY, Tannoury CA, et al. Cervical sprains, disc herniations, minor fractures, and other cervical injuries in the athlete. Clin Sports Med 2003;22:513-21.

  9. Browning R, Jackson JL, O'Malley PG. Cyclobenzaprine and back pain. A Meta analysis. Arch Intern Med 2001;161:1613-20.

  10. Basmajian JV. Cyclobenzaprine hydrochloride effect on skeletal muscle spasm in the lumbar region and neck: two double blind controlled clinical and laboratory studies. Arch Phys Med Rehabil 1978;59:58-63.

  11. Borenstein DG, Lacks S, Wiesel SW. Cyclobenzaprine and naproxen versus naproxen alone in the treatment of acute low back pain and muscle spasm. Clin Ther 1990;12:125-31.

  12. Turturro MA, Frater CR, D'Amico FJ. Cyclobenzaprine with ibuprofen versus ibuprofen alone in acute myofascial strain: a randomized, double blind clinical trial. Ann Emerg Med 2003;41:818-26.

  13. Hollander JE, Singer AJ. An innovative strategy for conducting clinical research: the Academic Associate Program. Acad Emerg Med 2002;9:134-7.

  14. Hoffman JR, Wolfson AB, Todd K, et al. Selective cervical spine radiography in blunt trauma: methodology of the National Emergency X Radiography Utilization Study (NEXUS). Ann Emerg Med 1998;32:461-9.

  15. Scott J, Huskisson EC. Graphic representation of pain. Pain 1976;2:175-84.

  16. Singer AJ, Kowalska A, Thode HC. Ability of patients to accurately recall the severity of acute painful events. Acad Emerg Med 2001;8:292-5.

  17. Hirayama J, Yamagata M, Ogata S, et al. Relationship between low back pain, muscle spasm and pressure pain thresholds in patients with lumbar disc herniation. Eur Spine J 2006;15:41-7.

  18. Bigos S, Bowyer O, Braen G, et al. Acute low back problems in adults. Clinical practice guideline, quick reference guide number 14. Rockville (MD): US Department of Health and Human Services, Public Health Service, Agency for Health Care Policy and Research; 1994.