Osteopathic Manual Treatment and Ultrasound Therapy for Chronic Low Back Pain: A Randomized Controlled Trial

John C. Licciardone; Dennis E. Minotti; Robert J. Gatchel; Cathleen M. Kearns; Karan P. Singh

doi:10.1370/afm.1468

Article Figures & Data

Figures

Tables
Additional Files

Download figure
Open in new tab
Download powerpoint
Figure 1
Allocation of patients to osteopathic manual treatment and ultrasound therapy interventions using a 2 × 2 factorial design.

View popup

Table 1

Baseline Patient Characteristics by Main Effects Group

Characteristic	OMT vs Sham OMT		UST vs Sham UST
Characteristic	OMT (n = 230)	Sham OMT (n = 225)	UST (n = 233)	Sham UST (n = 222)	Total (N = 455)
Age, median (IQR), y	41 (29-51)	40 (29-50)	38 (29-50)	43 (30-51)	41 (29-51)
Women, No. (%)	144 (63)	140 (62)	134 (58)	150 (68)	284 (62)
Completed college education, No. (%)	107 (47)	93 (41)	102 (44)	98 (44)	200 (44)
Employed full-time, No. (%)	110 (48)	105 (47)	114 (49)	101 (45)	215 (47)
Medically uninsured, No. (%)	86 (37)	77 (34)	79 (34)	84 (38)	163 (36)
Current smoker, No. (%)	61 (27)	58 (26)	61 (26)	58 (26)	119 (26)
Comorbid conditions, No. (%)
Hypertension	42 (18)	29 (13)	34 (15)	37 (17)	71 (16)
Diabetes mellitus	19 (8)	15 (7)	17 (7)	17 (8)	34 (7)
Osteoarthritis	17 (7)	16 (7)	15 (6)	18 (8)	33 (7)
Depression	44 (19)	46 (20)	41 (18)	49 (22)	90 (20)
Duration of chronic LBP >1 y, No. (%)	118 (51)	110 (49)	119 (51)	109 (49)	228 (50)
Previously hospitalized for LBP, No. (%)	13 (6)	8 (4)	9 (4)	12 (5)	21 (5)
Previously had surgery for LBP, No. (%)	5 (2)	5 (2)	3 (1)	7 (3)	10 (2)
VAS score for LBP (mm) median (IQR)^{^a}	44 (25-61)	45 (28-60)	44 (29-60)	44 (23 to 61)	44 (26-60)
Roland-Morris disability score, median (IQR)^{^b}	5 (3-9)	5 (3-10)	5 (3-10)	5 (3-9)	5 (3-9)
SF-36 general health score, median (IQR)^{^c}	67 (57-82)	72 (52-85)	72 (56-85)	67 (52-82)	72 (52-82)
Used drugs for LBP during past 4 wk, No. (%)
Nonprescription	115 (50)	107 (48)	119 (51)	103 (46)	222 (49)
Prescription	27 (12)	32 (14)	36 (15)	23 (10)	59 (13)

IQR = interquartile range; LBP = low back pain; OMT = osteopathic manual treatment; SF-36 = Medical Outcomes Study Short Form-36 Health Survey; UST = ultrasound therapy; VAS = visual analog scale.
↵a VAS (0-100 mm) used to measure LBP, with higher scores indicating more pain.
↵b Roland-Morris Disability Questionnaire (0-24 points) used to measure back-specific functioning, with higher scores indicating greater disability.
↵c SF-36 general health scale (0-100 points) used to measure generic health, with higher scores indicating better health.

View popup

Table 2

Treatment Effects for Osteopathic Manual Treatment and Ultrasound Therapy in Chronic Low Back Pain

LBP Reduction Threshold^a	OMT RR (95% CI)	UST RR (95% CI)
Intention-to-treat analysis (N = 455)
≥30%	1.38 (1.16-1.64)	1.02 (0.86-1.20)
≥50%	1.41 (1.13-1.76)	1.09 (0.88-1.35)
≥20 mm	1.47 (1.17-1.86)	1.01 (0.80-1.26)
≥40 mm	1.96 (1.18-3.24)	1.09 (0.68-1.75)
Per-protocol analysis (n = 362)
≥30%	1.42 (1.19-1.70)	1. 03 (0.87-1.23)
≥50%	1.48 (1.18-1.86)	1.11 (0.89-1.38)
≥20 mm	1.44 (1.13-1.85)	1.05 (0.83-1.34)
≥40 mm	2.08 (1.21-3.58)	1.01 (0.61-1.67)
Patients with missed treatments considered nonresponders (N = 455)
≥30%	1.38 (1.13-1.69)	0.97 (0.80-1.18)
≥50%	1.43 (1.12-1.83)	1.05 (0. 82-1.33)
≥20 mm	1.40 (1.07-1.82)	0.99 (0.77-1.28)
≥40 mm	2.01 (1.16-3.49)	0.95 (0.57-1.59)

Note: Response ratios are for active vs sham treatments. RR >1 potentially indicates some level of response to active treatment. Using the Cochrane Back Review Group guidelines,³³ significant treatment effects are further classified as small, RR <1.25; medium, RR = 1.25 to ≤2.0; or large, RR >2. Thus, table entries for OMT effects were all in the medium to large range. There was no evidence of any treatment effect with UST, as no table entry achieved significance.

IMMPACT = Initiative on Methods, Measurement, and Pain Assessment in Clinical Trials; LBP = low back pain; OMT = osteopathic manual treatment; RR = response ratio; UST = ultrasound therapy.
↵a IMMPACT benchmarks are ≥30% LBP reduction (moderate improvement) and ≥50% LBP reduction (substantial improvement).²⁴ Using absolute pain measures on a 100-mm visual analog scale, benchmarks are ≥20 mm LBP reduction (moderate improvement) and ≥40 mm LBP reduction (substantial improvement).

View popup

Table 3

Secondary Outcomes by Main Effects Group

Outcome	OMT vs Sham OMT			UST vs Sham UST
Outcome	OMT (n = 230)	Sham OMT (n = 225)	P Value	UST (n = 233)	Sham UST (n = 222)	P Value
RMDQ score, median (IQR)^a
Week 4	4 (2-8)	5 (2-9)	.32	4 (2-9)	5 (2-8)	.99
Week 8	3 (1-7)	3 (2-8)	.14	3 (1-8)	4 (1-7)	.76
Week 12	2 (1-6)	3 (1-7)	.07	3 (1-7)	3 (1-7)	.93
SF-36 GH score, median (IQR)^b
Week 4	71 (55-82)	72 (52-86)	.39	72 (54-87)	72 (52-82)	.73
Week 8	72 (57-85)	72 (52-85)	.61	72 (54-85)	72 (57-85)	.53
Week 12	72 (52-87)	72 (57-87)	.87	72 (52-87)	74 (54-87)	.66
Lost 1 or more work days in past 4 weeks because of LBP, % (95% CI)^c
Week 4	10 (4-16)	14 (7-21)	.41	16 (9-23)	7 (2-12)	.04
Week 8	6 (2-11)	19 (12-27)	.005	17 (10-24)	8 (3-14)	.054
Week 12	11 (5-17)	8 (3-13)	.41	13 (6-19)	6 (1-11)	.11
Very satisfied with back care, % (95% CI)^d
Week 4	52 (46-59)	34 (28-41)	<.001	41 (35-48)	45 (38-52)	.44
Week 8	61 (54-67)	39 (33-46)	<.001	49 (43-56)	51 (44-58)	.77
Week 12	66 (60-73)	43 (36-50)	<.001	55 (48-61)	55 (48-62)	.99
LBP co-treatment during study, % (95% CI)^e
Exercise programs	19 (14-24)	20 (14-25)	.82	20 (15-25)	18 (13-24)	.73
Lumbar supports	1 (0-3)	1 (0-2)	>.99	1 (0-2)	1 (0-3)	.68
Nonprescription drugs	46 (39-52)	45 (39-52)	.95	46 (40-53)	45 (38-51)	.71
Prescription drugs	13 (9-18)	20 (15-26)	.048	16 (11-21)	18 (13-23)	.54
CAM therapies	15 (11-20)	17 (12-22)	.63	16 (12-21)	16 (11-21)	.87
Physical therapy	11 (7-15)	8 (4-11)	.17	9 (5-13)	10 (6-14)	.74
Hospitalization	0 (0-0)	0 (0-1)	.49	0 (0-1)	0 (0-0)	>.99
Surgery	0 (0-1)	0 (0-0)	>.99	0 (0-0)	0 (0-1)	.49

CAM = complementary and alternative medicine; IQR = interquartile range; LBP = low back pain; OMT = osteopathic manual treatment; RMDQ = Roland-Morris Disability Questionnaire; SF-36 GH = Medical Outcomes Study Short Form-36 Health Survey general health scale; UST = ultrasound therapy.
↵a Higher scores on the RMDQ represent greater disability. P values at each endpoint based on the Mann-Whitney test; N = 455 for all endpoints.
↵b Higher scores on the SF-36 GH represent better health. P values at each endpoint based on the Mann-Whitney test; N = 455 for all endpoints.
↵c Work disability analyses limited to the 215 patients employed full-time at baseline; n = 207, 211, and 211 at the successive endpoints.
↵d n = 416, 426, and 429 at the successive endpoints.
↵e N = 455 for all chronic LBP co-treatments.

Additional Files

Figures
Tables

Supplemental Figure

Supplemental Figure 1. Patient flow diagram.

Files in this Data Supplement:

Supplemental data: Figure - PDF file, 1 page, 496 KB
The Article in Brief

Osteopathic Manual Treatment and Ultrasound Therapy for Chronic Low Back Pain: A Randomized Controlled Trial

John C. Licciardone , and colleagues

Background Low back pain is responsible for more than 20 million ambulatory medical care visits and $100 billion in costs annually in the United States. Research has not conclusively demonstrated efficacy of osteopathic manual treatment (OMT) in relieving low back pain. This study examines the efficacy of OMT and ultrasound therapy (UST) for treatment of chronic low back pain.

What This Study Found OMT is effective for short-term pain relief when used to complement other co-treatments for chronic low back pain. OMT met or exceeded the Cochrane Back Review Group criterion for a medium effect size for both moderate and substantial improvements in low back pain. Thus, low back pain reductions with OMT were statistically significant and clinically relevant. The less frequent use of prescription drugs for low back pain reported by OMT patients further corroborates the clinical relevance of these results. The OMT regimen was safe, parsimonious, and well accepted by patients as demonstrated by high levels of treatment adherence and satisfaction with back care. By contrast, UST was not efficacious in relieving chronic low back pain.

Implications

The authors suggest that these results may begin to explain why one-third of ambulatory, chronic problem visits for low back pain in the United States are provided by osteopathic physicians, and why they less frequently prescribe medications, such as nonsteroidal anti-inflammatory drugs, than allopathic physicians during such visits.