Research

Original Investigation

Naproxen With Cyclobenzaprine, Oxycodone/Acetaminophen,

or Placebo for Treating Acute Low Back Pain
A Randomized Clinical Trial
Benjamin W. Friedman, MD, MS; Andrew A. Dym, BS; Michelle Davitt, MD; Lynne Holden, MD;
Clemencia Solorzano, PharmD; David Esses, MD; Polly E. Bijur, PhD; E. John Gallagher, MD

Supplemental content at
IMPORTANCE Low back pain (LBP) is responsible for more than 2.5 million visits to US jama.com
emergency departments (EDs) annually. These patients are usually treated with CME Quiz at
nonsteroidal anti-inflammatory drugs, acetaminophen, opioids, or skeletal muscle relaxants, jamanetworkcme.com and
often in combination. CME Questions page 1640

OBJECTIVE To compare functional outcomes and pain at 1 week and 3 months after an ED
visit for acute LBP among patients randomized to a 10-day course of (1) naproxen + placebo;
(2) naproxen + cyclobenzaprine; or (3) naproxen + oxycodone/acetaminophen.

DESIGN, SETTING, AND PARTICIPANTS This randomized, double-blind, 3-group study was
conducted at one urban ED in the Bronx, New York City. Patients who presented with
nontraumatic, nonradicular LBP of 2 weeks’ duration or less were eligible for enrollment upon
ED discharge if they had a score greater than 5 on the Roland-Morris Disability Questionnaire
(RMDQ). The RMDQ is a 24-item questionnaire commonly used to measure LBP and related
functional impairment on which 0 indicates no functional impairment and 24 indicates
maximum impairment. Beginning in April 2012, a total of 2588 patients were approached for
enrollment. Of the 323 deemed eligible for participation, 107 were randomized to receive
placebo and 108 each to cyclobenzaprine and to oxycodone/acetaminophen. Follow-up was
completed in December 2014.

INTERVENTIONS All participants were given 20 tablets of naproxen, 500 mg, to be taken
twice a day. They were randomized to receive either 60 tablets of placebo; cyclobenzaprine,
5 mg; or oxycodone, 5 mg/acetaminophen, 325 mg. Participants were instructed to take 1 or 2
of these tablets every 8 hours, as needed for LBP. They also received a standardized
10-minute LBP educational session prior to discharge.

MAIN OUTCOMES AND MEASURES The primary outcome was improvement in RMDQ between
ED discharge and 1 week later.

RESULTS Demographic characteristics were comparable among the 3 groups. At baseline,

median RMDQ score in the placebo group was 20 (interquartile range [IQR],17-21), in the
cyclobenzaprine group 19 (IQR,17-21), and in the oxycodone/acetaminophen group 20 (IQR,17-
22). At 1-week follow-up, the mean RMDQ improvement was 9.8 in the placebo group, 10.1 in the
Author Affiliations: Department of
cyclobenzaprine group, and 11.1 in the oxycodone/acetaminophen group. Between-group Emergency Medicine, Montefiore
difference in mean RMDQ improvement for cyclobenzaprine vs placebo was 0.3 (98.3% CI, −2.6 Medical Center, Albert Einstein
to 3.2; P = .77), for oxycodone/acetaminophen vs placebo, 1.3 (98.3% CI, −1.5 to 4.1; P = .28), College of Medicine, Bronx, New York
(Friedman, Davitt, Holden, Esses,
and for oxycodone/acetaminophen vs cyclobenzaprine, 0.9 (98.3% CI, −2.1 to 3.9; P = .45).
Bijur, Gallagher); Medical College,
Albert Einstein College of Medicine,
CONCLUSIONS AND RELEVANCE Among patients with acute, nontraumatic, nonradicular LBP Bronx, New York (Dym); Pharmacy
presenting to the ED, adding cyclobenzaprine or oxycodone/acetaminophen to naproxen Department, Montefiore Medical
Center, Bronx, New York (Solorzano).
alone did not improve functional outcomes or pain at 1-week follow-up. These findings do not
Corresponding Author: Benjamin W.
support use of these additional medications in this setting.
Friedman, MD, MS, Department of
Emergency Medicine, Montefiore
TRIAL REGISTRATION clinicaltrials.gov Identifier: NCT01587274 Medical Center, Albert Einstein
College of Medicine, 111 E 210th St,
Bronx, NY 10467 (bwfriedmanmd
JAMA. 2015;314(15):1572-1580. doi:10.1001/jama.2015.13043 @gmail.com).

1572 (Reprinted) jama.com

Copyright 2015 American Medical Association. All rights reserved.

Downloaded From: http://jama.jamanetwork.com/ by a National University of Singapore User on 10/20/2015

 Pharmacological Treatment of Acute Low Back Pain Original Investigation Research

L
ow back pain is responsible for 2.4% of visits to US emer- management of acute LBP, defined as pain originating
gency departments (EDs) resulting in 2.7 million visits between the lower border of the scapulae and the upper glu-
annually.1 Pain outcomes for these patients are gener- teal folds, and received a diagnosis consistent with nontrau-
ally poor.2 One week after an ED visit in an unselected low back matic nonradicular, musculoskeletal LBP. Patients were
pain (LBP) population, 70% of patients reported persistent back required to have functionally impairing back pain, which we
pain–related functional impairment and 69% reported con- defined as a score of greater than 5 on the Roland-Morris
tinued analgesic use.2 Three months later, 48% reported func- Disability Questionnaire (RMDQ).14 The RMDQ is a 24-item
tional impairment and 46% reported persistent analgesic use, questionnaire commonly used to measure LBP and related
including 19% who required opioids.2 functional impairment (0 indicates no impairment; 24 indi-
A variety of evidence-based medications are available to treat cates maximum impairment). Patients were excluded for
LBP.3 Nonsteroidal anti-inflammatory drugs (NSAIDs) are more radicular pain, which we defined as pain radiating below the
efficacious than placebo.4 Skeletal muscle relaxants are effective gluteal folds, direct trauma to the back within the previous
for short-term pain relief and global efficacy.5 Opioids are com- month, pain duration for more than 2 weeks, or recent his-
monly used for moderate or severe acute LBP,6 although high- tory of greater than 1 LBP episode per month. We also
quality evidence supporting this practice is lacking. excluded patients who were pregnant or lactating, unavail-
Treatment of LBP with multiple concurrent medications able for follow-up, with allergy or contraindication to the
is common in the ED setting.7 In a study using data from a na- investigational medications, or had chronic opioid use cur-
tional sample gathered in 2002-2006, emergency physicians rently or in the past. Patients could only be enrolled once.
often prescribed NSAIDs, skeletal muscle relaxants, and opi-
oids in combination, that is, 26% of patients received an Interventions
NSAID combined with a skeletal muscle relaxant and another All patients received naproxen, twenty 500-mg tablets, taken
26% received an NSAID combined with an opioid.7 Sixteen per- as 1 every 12 hours. All patients also received 60 tablets of
cent of patients received all 3 classes of medication.7 Several one of the following investigational medications, to be taken
clinical trials have compared NSAIDS + skeletal muscle relax- as 1 or 2 tablets every 8 hours: (1) placebo; (2) cyclobenza-
ants to monotherapy with just one of these agents.8-12 These prine, 5 mg; or (3) oxycodone, 5 mg/acetaminophen, 325 mg.
trials have reported heterogeneous results. The combination In an effort to maximize effectiveness while minimizing
of opioids + NSAIDS has been insufficiently evaluated experi- adverse effects, patients were instructed to take 1 or 2 tablets
mentally in patients with acute LBP.13 of their randomly assigned medication, as needed, every 8
Given the pain and functional impairment that persists be- hours. If one tablet afforded sufficient relief, there was no
yond an ED visit for musculoskeletal LBP and the heteroge- need to take the second tablet. However, patients who had
neity in clinical care, we conducted a randomized clinical trial not experienced sufficient relief within 30 minutes of taking
(RCT) to determine whether a 10-day course of muscle relax- the first tablet were instructed to take the second. We dis-
ants or opioids combined with NSAIDs is more effective than pensed 60 capsules of the investigational medication to
NSAID monotherapy for the treatment of nontraumatic non- every patient, enough to last 10 days if the patient took 2 tab-
radicular low back pain. lets every 8 hours.
Research personnel provided each patient with a 10-
minute educational intervention based on information from
the National Library of Medicine.15 Research personnel re-
Methods viewed the topic with the patient in English or Spanish and an-
Overview swered questions. Each participant was informed that care-
In this RCT, patients were enrolled during an ED visit for LBP, fully chosen exercises and stretches may help alleviate pain
dispensed a 10-day supply of medication, and contacted and prevent future occurrences and that hot or cold packs,
by telephone at 7-day and 3-month follow-up. The Albert physical therapy, massage therapy, and acupuncture help some
Einstein College of Medicine institutional review board pro- patients.
vided ethical oversight. All participants provided written in-
formed consent. The study protocol is available online Randomization and Blinding
(Supplement 1). The pharmacist performed a stratified randomization in
blocks of 6 based on 2 sequences using a randomization
Study Setting plan generator.16 Patients were stratified based on results of
We conducted this study in the ED of Montefiore Medical Cen- the baseline RMDQ. The pharmacist masked the medication
ter, an urban teaching hospital with more than 100 000 adult by placing cyclobenzaprine, oxycodone/acetaminophen, or
visits annually. Salaried, trained, fluently bilingual (English and placebo into identical unmarked capsules, which were
Spanish) research associates staffed the ED 16 to 24 hours per then packed with small amounts of lactose and sealed. The
day, 7 days per week during the accrual period. pharmacist created research packets, each with 2 vials of
medication, one containing naproxen and the other contain-
Participant Selection ing the masked investigational medication. Research
Patients were considered for inclusion if they were adults packets were dispensed to study participants by research
aged 21 to 64 years who presented to the ED primarily for personnel.

jama.com (Reprinted) JAMA October 20, 2015 Volume 314, Number 15 1573

Copyright 2015 American Medical Association. All rights reserved.

Downloaded From: http://jama.jamanetwork.com/ by a National University of Singapore User on 10/20/2015

 Research Original Investigation Pharmacological Treatment of Acute Low Back Pain

Figure. Flow of Patients Through Acute Low Back Pain Trial

2588 Patients with low back pain

assessed for eligibility

2198 Excluded
1025 Low back pain duration >2 weeks
or more frequent than 1 × /mo
696 Age >65 years
288 Direct trauma
57 Medication contraindication
or allergy
53 Radicular pain
33 Medical cause of low back pain
19 Unable to consent
11 Previous enrollment
10 Chronic opioid use
6 Admitted to hospital
67 Refused participation

323 Randomized

107 Randomized to receive placebo 108 Randomized to receive 108 Randomized to receive oxycodone/
107 Received placebo as cyclobenzaprine acetaminophen
randomized 108 Received cyclobenzaprine 108 Received oxycodone/
as randomized acetaminophen as randomized

3 Lost to 7-day follow-up 5 Lost to 7-day follow-up 4 Lost to 7-day follow-up

11 Lost to 3-month follow-up a 10 Lost to 3-month follow-up a 8 Lost to 3-month follow-up a

107 Included in analysis b 108 Included in analysis b 108 Included in analysis b

a
Participants lost to follow-up were those we were unable to contact by telephone or mail.
b
Multiple imputation was used to account for missing data.

Outcome Measures outcomes; however, these were not described explicitly in

The primary outcome for this study was improvement on the the protocol (Supplement 1).
RMDQ between ED discharge and the 7-day telephone follow- Adverse events were ascertained by asking patients to re-
up. A 5-point improvement on this scale is generally consid- port any symptoms from the medications. We specifically
ered a clinically significant improvement.14 asked participants to describe whether or not the medica-
We assessed a number of exploratory outcomes at 1 tions irritated their stomach or made them tired or dizzy. For
week and 3 months after ED discharge. One week after ED these latter 3 symptoms, participants were asked to also use
discharge, we determined participants’ worst LBP during the the descriptions “a lot,” “a little,” or “none.”
previous 24 hours, using a 4-item ordinal scale (severe,
moderate, mild, or none); the frequency of any analgesic or Sample Size Calculation
LBP medication use during the previous 24 hours; the fre- We assumed a mean (SD) change in RMDQ between baseline
quency of LBP during the previous 24 hours using a 5-item and 1 week of 5.6 (10.0)2 and a minimum clinically important
scale (not at all, rarely, sometimes, usually, always); satisfac- difference on the RMDQ of 5.0.17 We incorporated a 2-tailed α
tion with treatment, as measured by response to the ques- of .02 to adjust for 3 pairwise comparisons and a β of .20. To
tion “The next time you have back pain, do you want to take account for the possibility of a nonparametric analysis, we
the same medications you’ve been taking this past week”; added a 15% surplus and an additional 10% to account for
following ED discharge, the day on which the participant those who were lost to follow-up, resulting in the need for
was able to return to work and resume all usual activities; 323 patients.
and the frequency of visits to any clinician. Three months
after ED discharge, we determined participants’ worst LBP Analysis
during the previous 72 hours, using the same ordinal scale as An intention-to-treat analysis was performed. A secondary pre-
previously described; the frequency of LBP during the previ- specified analysis, not described in the original protocol, in-
ous 72 hours using the previous scale; the frequency of use cluded only those patients who reported taking the assigned
of any LBP medication during the previous 72 hours; and the investigational medication more than once. The analysis of the
frequency of opioid use. We assessed the actual RMDQ score primary outcome consisted of 3 pairwise comparisons of the
at 1-week and 3-month follow-up as prespecified secondary change in RMDQ between baseline at ED discharge and 1 week

1574 JAMA October 20, 2015 Volume 314, Number 15 (Reprinted) jama.com

Copyright 2015 American Medical Association. All rights reserved.

Downloaded From: http://jama.jamanetwork.com/ by a National University of Singapore User on 10/20/2015

 Pharmacological Treatment of Acute Low Back Pain Original Investigation Research

Table 1. Baseline Characteristics

No. (%)a
Naproxen +
Naproxen + Naproxen + Oxycodone/
Placebo Cyclobenzaprine Acetaminophen
Variable (n = 107) (n = 108) (n = 108)
Age, mean (SD), y 39 (11) 38 (11) 39 (11) Abbreviations: ED, emergency
Sex department; IQR, interquartile range;
LBP, low back pain; RMDQ, Roland
Men 54 (50) 63 (58) 48 (44) Morris Disability Questionnaire.
Women 53 (50) 45 (42) 60 (56) a
Data are reported as No (%) unless
Educational levelb otherwise indicated.
b
Did not graduate 25 (24) 30 (28) 35 (33) Missing data for 4 participants.
from high school c
The RMDQ is a 24-item instrument
Some college 60 (57) 50 (47) 43 (41) measuring low back pain–related
functional impairment (0 indicates
Graduated from college 21 (20) 27 (25) 28 (26)
no functional impairment; 24
RMDQ score at time indicates maximum functional
of ED dischargec impairment).
Median (IQR) 20 (17-21) 19 (17-21) 20 (17-22) d
Denominators differ because of
Mean (SD) 18.7 (4.0) 18.4 (4.1) 18.9 (3.7) missing data.
e
Duration of LBP 48 (24-96) 48 (18-96) 72 (33-139) Patients were asked 2 screening
prior to ED presentation, questions from the Patient Health
median (IQR), h Questionnaire: “Before your back
Previous episodes pain began, how often were you
of LBP bothered by little pleasure or
Never 47 (44) 49 (45) 51 (47) interest in doing things?” and
“Before your back pain began, how
A few times 44 (41) 52 (48) 46 (43) often were you bothered by feeling
≥1/y 16 (15) 7 (7) 11 (10) down, depressed, or hopeless?”
Patients who responded to either
On-the-job injuryd 24/106 (23) 36/107 (34) 31/106 (29)
question “More than half the days”
Depression screen 8/106 (8) 2/108 (2) 5/108 (5) or “Nearly every day” were
positived,e considered screen positive.

later, reported with 98.3% CI. The significance threshold for benzaprine improved by 10.1 (98.3% CI, 7.9 to 12.3), and those
the primary outcome was .02. Exploratory outcomes were not randomized to naproxen + oxycodone/acetaminophen im-
adjusted for multiple comparisons. These are reported as be- proved by 11.1 (98.3% CI, 9.0 to 13.2). Between group differ-
tween-group differences with 95% CIs or difference between ences in mean RMDQ improvement were as follows: cyclo-
medians with 95% CIs. The number needed to treat (NNT) is benzaprine vs placebo was 0.3 (98.3% CI, −2.6 to 3.2; P = .77),
presented with a 95% CI when naproxen + active medication oxycodone/acetaminophen vs placebo was 1.3 (98.3% CI, −1.5
resulted in a statistically significant improvement in out- to 4.1; P = .28), and oxycodone/acetaminophen vs cycloben-
come compared with naproxen + placebo. The number needed zaprine was 0.9 (98.3% CI, −2.1 to 3.9; P = .45).
to harm (NNH) is presented with a 95% CI when naproxen + ac-
tive medication resulted in a statistically significant increase Exploratory Outcomes
in adverse events compared with naproxen + placebo. Mul- At 1-week follow-up, regardless of study group, more than 50%
tiple imputation was performed to account for missing data. of patients still required medication for LBP, and as shown in
IBM SPSS Statistics, version 21 was used for all analyses. Table 2, many patients reported moderate or severe, and fre-
quent pain. Despite these generally poor outcomes, more than
two-thirds of patients reported that they would want to re-
ceive the same medications during a subsequent ED visit for
Results acute LBP.
During a 30-month period beginning in April 2012, a total of More than 75% of participants randomized to receive
323 patients were enrolled (Figure). Follow-up was com- naproxen used it daily and nearly two-thirds used it twice
pleted in December 2014. Baseline characteristics were not dif- daily (Table 3). Fewer participants used the cyclobenzaprine,
ferent between the 3 groups (Table 1). Baseline scores on the oxycodone/acetaminophen, or placebo regularly; only one-
RMDQ were high in all 3 study groups, indicating substantial third of patients used the medication they were randomized
functional impairment at baseline. to receive more than once daily and nearly 40% used this
medication intermittently, only once, or not at all (Table 3).
Primary Outcome Use of additional health care resources was infrequent in the
At 1-week follow-up, patients randomized to receive 3 study groups. Most participants did not visit their primary
naproxen + placebo improved by a mean of 9.8 (98.3% CI, 7.9 care clinician or a complementary/alternative medicine prac-
to 11.7) on the RMDQ, those randomized to naproxen + cyclo- titioner prior to the 1-week follow-up (Table 3).

jama.com (Reprinted) JAMA October 20, 2015 Volume 314, Number 15 1575

Copyright 2015 American Medical Association. All rights reserved.

Downloaded From: http://jama.jamanetwork.com/ by a National University of Singapore User on 10/20/2015

 1576
Table 2. One-Week Outcomes Among All Study Participantsa

Difference, % (95% CI)

Naproxen + Naproxen + Naproxen + Oxycodone/ Oxycodone/ Cyclobenzaprine vs
Placebo Cyclobenzaprine Acetaminophen Cyclobenzaprine vs Acetaminophen vs Oxycodone/
Outcome Variable (n = 107) (n = 108) (n = 108) Placebo Placebo Acetaminophen
Primary Outcome
Improvement in RMDQ between ED visit and 1-week follow-up, 9.8 (7.9 to 11.7) 10.1 (7.9 to 12.3) 11.1 (9.0 to 13.2) 0.3 (−2.6 to 3.2)c 1.3 (−1.5 to 4.1)c 0.9 (−2.1 to 3.9)c
mean (98.3% CI)b
Exploratory Outcomes
Research Original Investigation

RMDQb
Mean (95% CI) score 8.9 (7.3 to 10.5) 8.2 (6.2 to 9.4) 7.8 (6.6 to 9.8) 0.7 (−1.6 to 3.0) 1.1 (−1.1 to 3.4) 0.4 (−1.2 to 2.7)
Median (IQR) score 7 (0 to 18) 4 (0 to 16) 5 (0 to 15)
Worst LBP during previous 24 h, No. (%)
Mild/none 58 (54) 65 (60) 70 (65)
6 (−7 to 19) 11 (−2 to 24) 5 (−8 to 18)
Moderate/severe 49 (46) 43 (40) 38 (35)
Frequency of LBP during previous 24 h, No. (%)

JAMA October 20, 2015 Volume 314, Number 15 (Reprinted)

Never/rarely 46 (43) 52 (48) 52 (48)
Sometimes 25 (23) 25 (23) 25 (23) 6 (−7 to 18)d 6 (−6 to 19)d 1 (−11 to 13)d
Frequently/always 37 (35) 31 (29) 30 (28)
Use of medication for LBP during the 24 h prior to one week
follow-up, No. (%)
No medications 39 (36) 46 (43) 49 (45)
6 (−7 to 19) 9 (−4 to 22) 3 (−10 to 16)
Took medications 68 (64) 62 (57) 59 (55)
Desires same medications during subsequent episode of LBP,
No. (%)e
Yes 72 (67) 81 (75) 79 (73)
No 26 (24) 18 (17) 22 (20) 8 (−4 to 20) 6 (−6 to 18) 2 (−10 to 14)
Not sure 9 (8) 9 (8) 7 (6)

Downloaded From: http://jama.jamanetwork.com/ by a National University of Singapore User on 10/20/2015

No. of days, median (IQR)
Return to usual activitiesf 5 (2 to >7) 4 (3 to >7) 4 (3 to >7) 0 (0 to 1) 0 (0 to 1) 0 (0 to 0)

Copyright 2015 American Medical Association. All rights reserved.

Return to workg 3 (2 to 4) 3 (2 to 4) 2 (1 to 4) 0 (−1 to 1) 0 (0 to 1) 0 (0 to 1)
d
Abbreviations: ED, emergency department; IQR, interquartile range; LBP, low back pain; RMDQ, Roland-Morris Never/rarely/sometimes vs frequently/always.
Disability Questionnaire. e
Participants were asked “The next time you have back pain, do you want to take the same medications you’ve
a
These data include imputed values for 12 patients lost to follow-up at 1-week follow-up. Multiple imputation been taking this past week?” Yes, no, or not sure.
models included the following variables: age, sex, investigational medication, and baseline RMDQ. Values were f
Patients who had not yet recovered at the time of the 1-week follow-up phone call were categorized as requiring
rounded to the nearest integer. more than 7 days before returning to usual activities.
b
The RMDQ is a 24-item instrument measuring low back pain–related functional impairment (0 indicates no g
Data include 273 employed participants; 50 participants (15.5% of study population) were unemployed.
functional impairment; 24 indicates maximum functional impairment).
c
Between-group difference for mean improvement on RMDQ.

jama.com
Pharmacological Treatment of Acute Low Back Pain
 Pharmacological Treatment of Acute Low Back Pain Original Investigation Research

Table 3. Use of Investigational Medication and Health Care Resources Within 1 Week of ED Discharge

No. (%) [95% CI]

Naproxen +
Naproxen + Naproxen + Oxycodone/
Outcome Placebo Cyclobenzaprine Acetaminophen
Frequency of naproxen use
No. of patients 101 103 104
>1/d 72 (72) [62-79] 67 (65) [55-74] 56 (54) [44-63]
1/d 13 (13) [8-21] 21 (20) [14-29] 26 (25) [18-34]
Sometimes 5 (5) [2-11] 5 (5) [2-11] 10 (10) [5-17]
Only once 6 (6) [3-13] 4 (4) [1-10] 7 (7) [3-13]
Never 5 (5) [2-11] 6 (6) [2-12] 5 (5) [2-11]
Frequency of placebo, cyclobenzaprine,
or oxycodone/acetaminophen use
No. of patients 104 102 102
>1/d 34 (33) [24-42] 32 (31) [23-41] 33 (32) [24-42]
1/d 32 (31) [23-40] 39 (38) [29-48] 21 (21) [14-30]
Sometimes 14 (14) [8-21] 13 (13) [7-21] 17 (17) [11-25]
Only once 8 (8) [4-15] 6 (6) [2-13] 10 (10) [5-17]
Never 16 (15) [10-24] 12 (12) [7-20] 21 (21) [14-30]
Health care resources used
No. of patients 104 103 103
No visits to any clinician 90 (87) [79-92] 87 (85) [76-90] 89 (86) [78-92] Abbreviation: ED, emergency
Subsequent ED visit 3 (3) [1-9] 1 (1) [0-6] 3 (3) [1-9] department.
a
Primary care visit 7 (7) [3-13] 11 (11) [6-18] 6 (6) [2-12] Includes a neurologist
or orthopedic clinician.
a
Visit to specialist physician 2 (2) [0-7] 0 (0) [0-4] 1 (1) [0-6] b
Includes a chiropractor
Complementary therapy visitb 4 (4) [1-10] 3 (3) [1-9] 6 (6) [2-12] or physical therapist.

Table 4. Adverse Medication Effectsa

No. (%) Difference, % (95% CI)

Naproxen +
Naproxen + Naproxen + Oxycodone/ Oxycodone/ Cyclobenzaprine vs
Placebo Cyclobenzaprine Acetaminophen Cyclobenzaprine vs Acetaminophen vs Oxycodone/
Adverse Event (n = 107) (n = 108) (n = 108) Placebo Placebo Acetaminophen
Any adverse event 22 (21) 36 (33) 43 (40) 13 (1 to 25) 19 (7 to 31) 6 (−6 to 19)
Drowsinessb 4 (4) 7 (7) 16 (15) 3 (−3 to 9) 11 (4 to 19) 8 (0 to 16)
Dizzinessb 3 (3) 3 (3) 16 (15) 0 (−4 to 4) 12 (5 to 19) 12 (5 to 19)
Stomach irritationb 5 (5) 7 (6) 7 (6) 2 (−4 to 8) 2 (−3 to 7) 0 (−6 to 7)
Nausea or vomiting 6 (6) 4 (4) 19 (18) 2 (−4 to 8) 12 (4 to 20) 14 (6 to 22)
a
These data include imputed values for 12 patients lost to follow-up at 1 week. not they experienced dizziness, drowsiness, and stomach irritation. They were
Multiple imputation models included the following variables: age, sex, asked to choose among the following options: “no,” “a little,” or “a lot.” Values
investigational medication, and baseline Roland-Morris Disability reported in this table indicate answers from participants whose symptoms
Questionnaire. were experienced a lot.
b
At the 7-day follow-up, study participants were asked specifically whether or

Adverse effects were more likely among patients ran- ized to oxycodone/acetaminophen were more likely than those
domized to receive oxycodone/acetaminophen than to pla- randomized to placebo to report pain levels of mild or none
cebo (difference, 19% [7% to 31%]; Table 4; number needed (difference, 18% [95% CI, 3% to 33%]; number needed to treat,
to harm, 5.3 [95% CI, 3 to 14]), and among patients random- 6 [95% CI, 3 to 37]).
ized to receive cyclobenzaprine vs placebo (difference, Three months after the ED visit, most patients had recov-
13% [1% to 25%]; Table 4; number needed to harm, 7.8 [95% ered, although nearly one-fourth in each study group still re-
CI, 4 to 129]). Other than the adverse effects listed in ported moderate or severe LBP and use of medication for LBP
Table 4, none occurred in more than 3 participants in any (Table 5). Opioid use for treating LBP was reported by 2.3% (95%
study group. CI, 0.8 to 5.3%) of participants.
Among the patients who used the cyclobenzaprine, oxyco- Additional data for the exploratory outcomes of pain in-
done/acetaminophen, or placebo investigational medication tensity at one week follow-up and resumption of usual activi-
more than once, there was no significant difference in the pri- ties at three month follow-up are reported in eTable 2 in
mary outcome (eTable 1 in Supplement 2). Patients random- Supplement 2.

jama.com (Reprinted) JAMA October 20, 2015 Volume 314, Number 15 1577

Copyright 2015 American Medical Association. All rights reserved.

Downloaded From: http://jama.jamanetwork.com/ by a National University of Singapore User on 10/20/2015

 Research Original Investigation Pharmacological Treatment of Acute Low Back Pain

Table 5. Three-Month Outcomesa

Naproxen + Difference, % (95% CI)

Naproxen + Naproxen + Oxycodone/ Oxycodone/ Cyclobenzaprine vs
Placebo Cyclobenzaprine Acetaminophen Cyclobenzaprine vs Acetaminophen vs Oxycodone/
Outcome Variable (n = 107) (n = 108) (n = 108) Placebo Placebo Acetaminophen
RMDQb
Mean (95% CI) score 3.8 (2.6 to 5.1) 4.5 (3.0 to 5.9) 4.6 (3.2 to 6.1) 0.6 (−1.3 to 2.6) 0.8 (−1.1 to 2.7) 0.2 (−1.9 to 2.2)
Median (IQR) score 0 (0 to 3) 0 (0 to 5) 0 (0 to 8)
No. (%)
Worst LBP during previous 72 h
Mild/none 79 (74) 81 (75) 87 (81)
1 (−11 to 13) 7 (−4 to 18) 6 (−6 to 17)
Moderate/severe 28 (26) 27 (25) 21 (19)
Frequency of LBP during previous 72 h
Never/rarely 78 (73) 80 (74) 79 (73)
Sometimes 10 (9) 16 (15) 11 (10) 7 (−3 to 16)c 1 (−9 to 11)c 6 (−4 to 15)c
Frequently/always 19 (18) 12 (11) 18 (17)
Use of medication for LBP within 72 h
No meds 79 (74) 82 (76) 88 (81)
2 (−10 to 14) 8 (−3 to 19) 6 (−5 to 16)
Took meds 28 (26) 26 (24) 20 (19)
Opioid use within 72 h 3 (3) 1 (1) 2 (2) 2 (−2 to 5) 1 (−3 to 5) 1 (−2 to 4)
b
Abbreviations: IQR, interquartile range; LBP, low back pain; RMDQ, Roland The RMDQ is a 24-item instrument measuring low back pain–related
Morris Disability Questionnaire. functional impairment; 0 indicates no functional impairment; 24 indicates
a
These data include imputed values for 29 patients lost to follow-up at 3-month maximum functional impairment.
c
follow-up. Multiple imputation models included the following variables: age, Differences were based on never/rarely/sometimes responses vs
sex, investigational medication, baseline RMDQ, 7-d RMDQ. frequently/always.

acute LBP. Available data do not support the superiority of opi-

Discussion oids over NSAIDs.19 In this study, oxycodone/acetaminophen +
naproxen was not better than placebo + naproxen. We identi-
In this RCT studying ED patients with acute, nontraumatic, fied a difference in pain outcomes among participants who
nonradicular musculoskeletal LBP, neither naproxen com- took oxycodone/acetaminophen more than once when com-
bined with oxycodone/acetaminophen nor naproxen com- pared with those who took placebo more than once. How-
bined with cyclobenzaprine provided better pain relief or bet- ever, this was one of multiple exploratory comparisons.
ter improvement in functional outcomes than naproxen Furthermore, the magnitude of benefit was modest, with a
combined with placebo. Measures of pain, functional impair- number needed to treat of nearly 6 for moderate or severe
ment, and use of health care resources were not different be- pain, which is balanced by a number needed to harm of
tween the study groups at 7 days or at 3 months after the ED nearly 5. Although we cannot exclude the possibility of a
visit. Regardless of allocation, nearly two-thirds of patients modest benefit of the opioid combination in a select sub-
demonstrated clinically significant improvement in LBP and group of patients, our data do not support providing
function 1 week later. However, 40% of the cohort reported oxycodone/acetaminophen in addition to naproxen for all
moderate or severe pain, half reported functionally impair- ED patients with acute LBP.
ing LBP, and nearly 60% were still using medication for their Many ED patients had already taken NSAIDs for LBP
LBP 1 week later. By 3-month follow-up, nearly one-fourth of before arriving to the ED. Some patients may have taken
the cohort reported moderate or severe pain and use of medi- insufficient doses at incorrect intervals and could be helped
cations for LBP. Three months after the ED visit, regardless of by optimization of their NSAID regimen. However, for
study group, opioid use for LBP was uncommon, with fewer patients who have already optimized their NSAID regimen,
than 3% of patients reporting use of an opioid within the pre- there are no additional evidence-based medical therapies
vious 72 hours. available. Our results show that that adding cyclobenzaprine
Our results are similar to other studies of NSAIDs com- or oxycodone/acetaminophen to naproxen does not improve
bined with cyclobenzaprine8,10,11,18 conducted in a variety of pain at 7-day or 3-month follow-up. It is also true that
settings, including an ED and primary care and specialty clin- corticosteroids20 and acetaminophen21 are not beneficial for
ics. Despite the fact that both NSAIDS and cyclobenzaprine are patients with nonradicular LBP. Benzodiazepines are some-
efficacious when administered as monotherapy,5,19 the bulk times used to treat LBP, although whether they are effective
of the data, including the findings in this study, suggest com- is unclear. 22 Remaining active leads to better outcomes
bination therapy is not better than monotherapy. than bed rest. 2 3 Complementary therapies, including
We were unable to find high-quality published data that acupuncture, 24 yoga, 25 and massage, 26 have been inad-
evaluated the efficacy of opioids combined with NSAIDs for equately studied in an acute LBP population. Data on exer-

1578 JAMA October 20, 2015 Volume 314, Number 15 (Reprinted) jama.com

Copyright 2015 American Medical Association. All rights reserved.

Downloaded From: http://jama.jamanetwork.com/ by a National University of Singapore User on 10/20/2015

 Pharmacological Treatment of Acute Low Back Pain Original Investigation Research

cise therapy for acute LBP is conflicting.27 Spinal manipula- LBP outcomes.36 Our study provides evidence against the
tion is unlikely to benefit ED patients with acute LBP whose use of opioids for acute LBP because of lack of benefit and
symptoms are well-managed medically.28 Emergency physi- increased frequency of adverse effects. However, opioids
cians should inform their patients that passage of time is were not associated with higher rates of functional impair-
likely to bring improvement and eventual relief to most ment, more frequent visits to the ED, or an increased pro-
patients. pensity for continued opioid use.
Our results are consistent with other reports of outcomes Our study has limitations. First, this study was con-
after acute-onset LBP.29-34 In general, most patients with acute- ducted in an urban ED that served a socioeconomically de-
onset LBP report persistent symptoms 1 week later. By 3 pressed population. Because back pain outcomes may be as-
months, however, most symptoms have improved. Risk fac- sociated with socioeconomic variables such as access to
tors for poor long-term LBP outcomes consist of complicated treatment, our results can most appropriately be generalized
LBP histories and radicular symptoms.35 In our study, we se- to EDs that serve similar patient populations.
lected patients at low risk of poor outcome by excluding those Second, we reported a large number of related outcomes.
with chronic LBP, radicular symptoms, or chronic use of opi- This approach may lead to uncertainty with regard to inter-
oids. Despite selecting for these low-risk patients, more than pretation of the data when some of the outcomes result in a
20% of our cohort, regardless of study group, reported poor statistically significant benefit and others do not. In this study,
outcomes at 3 months after the ED visit. we saw no difference in outcomes between those random-
We aimed to maximize medication use by instructing pa- ized to receive naproxen + placebo vs those randomized to re-
tients to choose whether to take 1 or 2 tablets of the investi- ceive naproxen + oxycodone/acetaminophen. However, among
gational medication at each dosing, thereby giving the pa- patients who used the investigational medication more than
tient the ability to titrate efficacy against adverse effects. once, fewer patients who used oxycodone/acetaminophen re-
Infrequent use of the study medication is both a limitation and ported moderate or severe pain. These latter findings must be
strength of this study—it is possible that standing doses of interpreted cautiously because of the large number of analy-
oxycodone/acetaminophen or cyclobenzaprine may have ses we performed.
treated the pain and functional impairment more effectively. Third, we did not evaluate the adequacy of patient blind-
However, we chose this study design because it more closely ing. Thus, we do not know whether patients’ assumptions
reflects the reality of clinical practice. about the investigational medication they were receiving in-
We dispensed 60 tablets of oxycodone/acetaminophen fluenced their self-reports of pain and functional outcomes.
to one-third of study participants—a substantial number of Fourth, we did not determine whether participants were using
opioid tablets—with the goal of not limiting potential benefit NSAIDs at the time of enrollment, thus limiting this study’s gen-
because of insufficient dosing. This liberal approach to opi- eralizability.
oids is at odds with recent clinical practice guidelines.36
Current recommendations state that when treating LBP, the
lowest possible dose of opioids should be prescribed for the
shortest amount of time. During the last 2 decades, there
Conclusions
has been increasing focus on achieving adequate pain con- Among patients with acute, nontraumatic, nonradicular LBP
trol, with the goal of quick and effective relief of pain. 37 presenting to an ED, adding cyclobenzaprine or oxycodone/
However, permissive use of opioids may be harmful. Using acetaminophen to naproxen alone did not improve func-
correlative data, others have linked prescriptions for opioids tional outcomes or pain at 7 days. These findings do not sup-
to overdose deaths and use of opioids in general to worse port the use of these additional medications in this setting.

ARTICLE INFORMATION REFERENCES 4. Roelofs PD, Deyo RA, Koes BW, Scholten RJ,
Author Contributions: Dr Friedman had full access 1. Friedman BW, Chilstrom M, Bijur PE, van Tulder MW. Nonsteroidal anti-inflammatory
to all of the data in the study and takes Gallagher EJ. Diagnostic testing and treatment of drugs for low back pain: an updated Cochrane
responsibility for the integrity of the data and the low back pain in United States emergency review. Spine (Phila Pa 1976). 2008;33(16):1766-1774.
accuracy of the data analysis. departments: a national perspective. Spine (Phila 5. van Tulder MW, Touray T, Furlan AD, Solway S,
Study concept and design: Friedman, Davitt, Pa 1976). 1976;35(24):E1406-E1411. Bouter LM; Cochrane Back Review Group. Muscle
Solorzano, Esses, Bijur, Gallagher. 2. Friedman BW, O'Mahony S, Mulvey L, et al relaxants for nonspecific low back pain:
Acquisition, analysis, or interpretation of data: One-week and 3-month outcomes after an a systematic review within the framework of the
Friedman, Dym, Holden, Esses, Bijur, Gallagher. emergency department visit for undifferentiated cochrane collaboration. Spine (Phila Pa 1976). 2003;
Drafting of the manuscript: Friedman. musculoskeletal low back pain. Ann Emerg Med. 28(17):1978-1992.
Critical revision of the manuscript for important 2012;59(2):128-133.e3. 6. Cherkin DC, Wheeler KJ, Barlow W, Deyo RA.
intellectual content: Friedman, Dym, Davitt, Holden, Medication use for low back pain in primary care.
Solorzano, Esses, Bijur, Gallagher. 3. Chou R, Qaseem A, Snow V, et al; Clinical Efficacy
Assessment Subcommittee of the American College Spine (Phila Pa 1976). 1998;23(5):607-614.
Statistical analysis: Friedman, Bijur.
Study supervision: Friedman. of Physicians; American College of Physicians; 7. Friedman BW, Chilstrom M, Bijur PE,
American Pain Society Low Back Pain Guidelines Gallagher EJ. Diagnostic testing and treatment of
Conflict of Interest Disclosures: All authors have Panel. Diagnosis and treatment of low back pain: low back pain in United States emergency
completed and submitted the ICMJE Form for a joint clinical practice guideline from the American departments: a national perspective. Spine (Phila
Disclosure of Potential Conflicts of Interest and College of Physicians and the American Pain Pa 1976). 2010;35(24):E1406-E1411.
none were reported. Society. Ann Intern Med. 2007;147(7):478-491.

jama.com (Reprinted) JAMA October 20, 2015 Volume 314, Number 15 1579

Copyright 2015 American Medical Association. All rights reserved.

Downloaded From: http://jama.jamanetwork.com/ by a National University of Singapore User on 10/20/2015

 Research Original Investigation Pharmacological Treatment of Acute Low Back Pain

8. Childers MK, Borenstein D, Brown RL, et al. 18. Turturro MA, Frater CR, D’Amico FJ. therapy in acute low back pain. Ann Rheum Dis.
Low-dose cyclobenzaprine versus combination Cyclobenzaprine with ibuprofen versus ibuprofen 2009;68(9):1420-1427.
therapy with ibuprofen for acute neck or back pain alone in acute myofascial strain: a randomized, 29. Cherkin DC, Deyo RA, Street JH, Hunt M,
with muscle spasm: a randomized trial. Curr Med double-blind clinical trial. Ann Emerg Med. 2003;41 Barlow W. Pitfalls of patient education: limited
Res Opin. 2005;21(9):1485-1493. (6):818-826. success of a program for back pain in primary care.
9. Berry H, Hutchinson DR. Tizanidine and 19. Roelofs PD, Deyo RA, Koes BW, Scholten RJ, Spine (Phila Pa 1976). 1996;21(3):345-355.
ibuprofen in acute low-back pain: results of a van Tulder MW. Non-steroidal anti-inflammatory 30. Coste J, Delecoeuillerie G, Cohen de Lara A,
double-blind multicentre study in general practice. drugs for low back pain. Cochrane Database Syst Rev. Le Parc JM, Paolaggi JB. Clinical course and
J Int Med Res. 1988;16(2):83-91. 2008;(1):CD000396. prognostic factors in acute low back pain: an
10. Basmajian JV. Acute back pain and spasm: 20. Friedman BW, Holden L, Esses D, et al. inception cohort study in primary care practice. BMJ.
a controlled multicenter trial of combined analgesic Parenteral corticosteroids for Emergency 1994;308(6928):577-580.
and antispasm agents. Spine (Phila Pa 1976). 1989; Department patients with non-radicular low back 31. Grotle M, Brox JI, Veierød MB, Glomsrød B,
14(4):438-439. pain. J Emerg Med. 2006;31(4):365-370. Lønn JH, Vøllestad NK. Clinical course and
11. Borenstein DG, Lacks S, Wiesel SW. 21. Williams CM, Maher CG, Latimer J, et al. Efficacy prognostic factors in acute low back pain: patients
Cyclobenzaprine and naproxen versus naproxen of paracetamol for acute low-back pain: consulting primary care for the first time. Spine
alone in the treatment of acute low back pain and a double-blind, randomised controlled trial. Lancet. (Phila Pa 1976). 2005;30(8):976-982.
muscle spasm. Clin Ther. 1990;12(2):125-131. 2014;384(9954):1586-1596. 32. Henschke N, Maher CG, Refshauge KM, et al.
12. Pareek A, Chandurkar N, Chandanwale AS, 22. Chou R. Pharmacological management of low Prognosis in patients with recent onset low back
Ambade R, Gupta A, Bartakke G. Aceclofenac- back pain. Drugs. 2010;70(4):387-402. pain in Australian primary care: inception cohort
tizanidine in the treatment of acute low back pain: 23. Hagen KB, Jamtvedt G, Hilde G, Winnem MF. study. BMJ. 2008;337:a171.
a double-blind, double-dummy, randomized, The updated cochrane review of bed rest for low 33. Roland MO. The natural history of back pain.
multicentric, comparative study against back pain and sciatica. Spine (Phila Pa 1976). 2005; Practitioner. 1983;227(1381):1119-1122.
aceclofenac alone. Eur Spine J. 2009;18(12):1836- 30(5):542-546.
1842. 34. Thomas E, Silman AJ, Croft PR, Papageorgiou
24. Furlan AD, van Tulder M, Cherkin D, et al. AC, Jayson MI, Macfarlane GJ. Predicting who
13. Deyo RA, Von Korff M, Duhrkoop D. Opioids for Acupuncture and dry-needling for low back pain: an develops chronic low back pain in primary care:
low back pain. BMJ. 2015;350:g6380. updated systematic review within the framework of a prospective study. BMJ. 1999;318(7199):1662-1667.
14. Roland M, Fairbank J. The Roland-Morris the cochrane collaboration. Spine (Phila Pa 1976). 35. Friedman BW, Mulvey L, Davitt M, et al.
Disability Questionnaire and the Oswestry Disability 2005;30(8):944-963. Predicting 7-day and 3-month functional outcomes
Questionnaire. Spine (Phila Pa 1976). 2000;25(24): 25. Cramer H, Lauche R, Haller H, Dobos G. after an ED visit for acute nontraumatic low back
3115-3124. A systematic review and meta-analysis of yoga for pain. Am J Emerg Med. 2012;30(9):1852-1859.
15. National Institute of Arthritis and low back pain. Clin J Pain. 2013;29(5):450-460. 36. Cantrill SV, Brown MD, Carlisle RJ, et al;
Musculoskeletal and Skin Diseases. What is back 26. Furlan AD, Imamura M, Dryden T, Irvin E. American College of Emergency Physicians Opioid
pain? Fast facts: an easy-to-read series of Massage for low back pain: an updated systematic Guideline Writing Panel. Clinical policy: critical
publications for the public. http://www.niams.nih review within the framework of the Cochrane Back issues in the prescribing of opioids for adult
.gov/Health_Info/Back_Pain/default.asp. Accessed Review Group. Spine (Phila Pa 1976). 2009;34(16): patients in the emergency department. Ann Emerg
September 28, 2015. 1669-1684. Med. 2012;60(4):499-525.
16. Randomization plan generator. http://www 27. Hayden JA, van Tulder MW, Malmivaara AV, 37. Fosnocht DE, Swanson ER, Barton ED.
.randomization.com. Accessed September 28, 2015. Koes BW. Meta-analysis: exercise therapy for Changing attitudes about pain and pain control in
17. Ostelo RW, Deyo RA, Stratford P, et al. nonspecific low back pain. Ann Intern Med. 2005; emergency medicine. Emerg Med Clin North Am.
Interpreting change scores for pain and functional 142(9):765-775. 2005;23(2):297-306.
status in low back pain: towards international 28. Jüni P, Battaglia M, Nüesch E, et al.
consensus regarding minimal important change. A randomised controlled trial of spinal manipulative
Spine (Phila Pa 1976). 2008;33(1):90-94.

1580 JAMA October 20, 2015 Volume 314, Number 15 (Reprinted) jama.com

Copyright 2015 American Medical Association. All rights reserved.

Downloaded From: http://jama.jamanetwork.com/ by a National University of Singapore User on 10/20/2015