ORIGINAL ARTICLE Multiﬁdus Atrophy Is Localized and Bilateral in Active Persons With Chronic Unilateral Low Back Pain George J. Beneck, PhD, PT, Kornelia Kulig, PhD, PT ABSTRACT. Beneck GJ, Kulig K. Multiﬁdus atrophy is localized and bilateral in active persons with chronic unilateral low back pain. Arch Phys Med Rehabil 2012;93:300-6. Objective: To compare the lumbar multiﬁdi muscle volume devoid of fat local to the site of pain in persons with and without chronic unilateral lower lumbar pain. Design: Prospective cross-sectional design. Setting: University biokinesiology laboratory. Participants: Active individuals (n=14) with chronic unilat- eral lower lumbar pain (1y) were matched for age, height, weight, and activity level with healthy individuals (n=14). Individuals with back pain had minimal disability (Oswestry Disability Index [mean  SD], 14.9%6.3%) at the time of testing. Interventions: Not applicable. Main Outcome Measures: Multiﬁdus and erector spinae mus- cle volumes at the L5-S1 levels, multiﬁdus muscle volumes at the L4 and S2-3 levels. Results: Average multiﬁdus volume was diminished by 18.1% between groups (P=.026) only at the L5-S1 levels. There was no difference between painful and pain-free sides. There were no volume differences between groups above L5, below S1, or in erector spinae at the L5-S1 levels. Conclusions: The results of this study indicate that despite a low level of disability and an activity level similar to that of matched control subjects, considerable localized, bilateral mul- tiﬁdus atrophy is present. Such impaired size of the multiﬁdus will likely reduce its capacity to control intersegmental motion, thus increasing the susceptibility to further injury. Unlike acute unilateral low back pain (LBP), muscle size is reduced bilat- erally in persons with chronic unilateral LBP. Key Words: Atrophy; Low back pain; Morphology; Reha- bilitation. © 2012 by the American Congress of Rehabilitation Medicine B ETWEEN 70% AND 80% OF adults experience at least 1 episode of low back pain (LBP) during their lifetime. 1 While most recover within 1 to 3 months, 2 40% experience a second episode within 6 months. 3 Despite various treatments, many continue to have repeated episodes of LBP several years later. 4 As a result, disability and health care costs are dispro- portionately higher for individuals with recurrent and chronic LBP than those experiencing their ﬁrst episode. 5,6 Laboratory, preclinical, and trial research point to the im- portance of lumbar muscle performance in spinal health. 7-11 As control and dynamic stability of the trunk are the unique performance demands on lumbar paraspinal muscles, muscle morphology is one of the quantiﬁable metrics contributing to muscle performance. 12 Impairment of the muscular stabilizing system of the lumbar spine has been related to chronic pain and repeated episodes of LBP. 13,14 The lumbar multiﬁdus, an im- portant component of this muscular stabilizing system, is con- sidered to be a vital stabilizer of the functional spinal units of the lumbar spine. In contrast to other lumbar paraspinal mus- cles, the physiologic cross-sectional area (CSA) of the lumbar multiﬁdus is more than twice that of either the longissimus thoracis or the iliocostalis lumborum. 15 In vitro studies 16,17 indicate that the multiﬁdus may control intervertebral motion by stiffening the spine. While each of the local paraspinal muscles contributes to spinal stability, the multiﬁdus alone is responsible for more than two thirds of the muscular stiffness in the sagittal plane. 17 Multiﬁdus atrophy has been reported in individuals with acute LBP, 18,19 chronic LBP, 20 and lumbar disk hernia- tions 21-24 and is associated with poor functional outcomes after disk surgery. 25,26 In addition to atrophy, fat inﬁltration in the lumbar extensors is a common ﬁnding in persons with chronic LBP. 27,28 Fat inﬁltration in the lumbar extensors is positively associated with LBP, 29 poorer physical performance, 30 and more disability. 28 With pain durations of less than 4 months, multiﬁdus atrophy is speciﬁc to the side and segmental region of pain. 18,31 However, it is unclear whether such morphologic atrophy, either unilateral or bilateral, is maintained in active persons with much longer durations of chronic LBP. There are several limitations in the previous methods used to measure multiﬁdus size. With the use of ultrasound, muscle borders are frequently difﬁcult to deﬁne, and fatty inﬁltrations cannot be distinguished from muscle. Previous studies used a limited number of images representing selected lumbar regions to characterize this multisegmental muscle. Thus, volumetric measurements of muscle, devoid of fat inﬁltration, would pro- vide a more comprehensive description of the multiﬁdus mor- phology associated with LBP. Therefore, the purpose of the study was to compare the volume of the multiﬁdus in persons with and without chronic unilateral lower lumbar pain. We hypothesized that the multiﬁdus volume would be reduced From the Department of Physical Therapy, California State University Long Beach, Long Beach, CA (Beneck); and Division of Biokinesiology and Physical Therapy, University of Southern California, Los Angeles, CA (Kulig). Presented to the American Physical Therapy Association, February 19, 2010, San Diego, CA; and to the California Physical Therapy Association, October 2, 2010, Oakland, CA. Supported by the Health Research Association (HRA) of the University of South- ern California (HRA Seed grant; project no. 227500) and by the Promotion of Doctoral Studies Scholarship through the Foundation for Physical Therapy. No commercial party having a direct ﬁnancial interest in the results of the research supporting this article has or will confer a beneﬁt on the authors or on any organi- zation with which the authors are associated. Correspondence to George J. Beneck, PhD, PT, Dept of Physical Therapy, ET 122, California State University, Long Beach, 1250 Bellﬂower Blvd, Long Beach, CA 90840, e-mail: gbeneck@csulb.edu. Reprints are not available from the authors. 0003-9993/12/9302-00628$36.00/0 doi:10.1016/j.apmr.2011.09.017 List of Abbreviations ANOVA analysis of variance CSA cross-sectional area LBP low back pain MR magnetic resonance 300 Arch Phys Med Rehabil Vol 93, February 2012