Secondary Logo

Journal Logo

Basic Science

Scientific Basis of Minimally Invasive Spine Surgery

Prevention of Multifidus Muscle Injury During Posterior Lumbar Surgery

Kim, Choll W. MD, PhD*†

Author Information
doi: 10.1097/BRS.0b013e3182022d32
  • Free

The treatment goals of minimally invasive spine (MIS) surgery are analogous to that of traditional open surgery. Goals of MIS surgery include decompression in cases where there is symptomatic nerve compression, fusion and/or instrumentation in cases when there is instability, and realignment in cases when there is clinically relevant deformity. More specifically, what distinguishes MIS surgery from traditional open surgery is its emphasis on the following: (1) decreasing muscle crush injuries during retraction; and (2) avoiding disruption of the osseotendinous complex of the paraspinal muscles, especially the multifidus attachments to the spinous process and superior articular processes. This is accomplished through the use of specialized instruments and refined surgical techniques where multiple smaller incisions are used to exploit known anatomic neurovascular and bone-tendon-muscle planes. Together, the goals of MIS are to maintain the dynamic stability of the spine while nonetheless accomplishing the intended goals of surgery.

Anatomy of the Posterior Paraspinal Muscles

The most surgically relevant posterior paraspinal muscles are composed of 3 major muscles: (1) multifidus, (2) longissimus, and (3) iliocostalis (Figure 1). The multifidus muscle is in intimate contact with a group of smaller muscles that lie between the spinous processes (interspinales), lamina (short rotators), and the transverse processes (intertransversarii).1 The multifidus, longissimus, and iliocostalis muscles run along the thoracolumbar spine and attach caudally to the sacrum, the sacroiliac joint, and the iliac wing. The multifidus is the most medial of the major posterior paraspinal muscles and is the largest muscle that spans the lumbosacral junction. It is believed to be the major posterior stabilizing muscle of the spine.2–4 Compared with other paraspinal muscles, the multifidus muscle is uniquely designed to be short and stout. It has a large physiologic cross sectional area (PCSA) but short fiber lengths. This architectural design allows the multifidus to create large forces for relatively short distances (Figure 2A).4 Furthermore, the multifidus sarcomere length is positioned on the ascending portion of the length-tension curve (Figure 2B). When our posture changes from standing erect to bending forward, the multifidus is able to produce more force as the spine flexes forward. This serves to protect the spine at its most vulnerable position. The unique design of the multifidus muscles suggests that the other paraspinal muscles will not be able to compensate well in its function as a key stabilizer of the lumbar spine.

Figure 1
Figure 1:
MR cross-sectional image through L4–L5 disc space showing the multifidus (M), iliocostalis (IL), longissimus (LO), quadratus lumborum (QL), intertransverserii (IT), and psoas muscles.
Figure 2
Figure 2:
A, Scatter plot of physiological cross sectional area (PCSA) versus fiber length. As PCSA is proportional to muscle force and fiber length is proportional to muscle excursion, this type of plot illustrates the functional design of a muscle. These data illustrate that the multifidus has the largest force generating capacity in the lumbar spine and is designed for stability. B, Sarcomere length operating range of the multifidus plotted on the human skeletal muscle sarcomere length tension curve (black line). These data demonstrate that the mult ifidus muscle operates on the ascending limb of the length tension curve and becomes intrinsically stronger as the spine in flexed (arrow). Schematic sarcomeres are shown on the ascending and descending limb to scale, based on the quantification of actin and myosin filament lengths. Reprinted with permission from Spine.41 Copyright 2009, Lippincott Williams & Wilkins.

The osseotendinous anatomy of the multifidus muscle is noteworthy. Although the tendon origin is from a single level at the spinous process, the caudal insertion sites are formed by multiple separate bands, each having its own distal attachments into the mammillary processes of the caudal vertebrae 2 to 5 levels below and downward through each vertebra to the sacrum. For example, fibers from the L1 band (which has a single insertion site at the L1 spinous process) insert into the mammillary processes of the L3, L4, and L5 vertebrae, to the dorsal part of S1, and to the posterior superior iliac spine. The multifidus is the only muscle that is attached both to the posterior parts of the L5 and S1 vertebrae and is, therefore, the sole posterior stabilizer that both originates and inserts to this segment.5

On the basis of the multifidus muscle anatomy, biomechanical analyses show that it produces posterior sagittal rotation of the vertebra, which opposes a counter rotation generated by the abdominal muscles. The multifidus can further increase lumbar spine stability through a “bowstring” mechanism in which the muscle, positioned posterior to the lumbar lordosis, produces compressive forces on the vertebrae interposed between its attachments.6

Although the multifidus is primarily responsible for dynamic stability of the lumbar spine, the erector spinae muscles comprised of the longissimus, the iliocostalis, and the spinalis (in the thoracic area) are mainly designed for trunk motion.5,7 The longissimus is positioned medially and arises from the transverse and accessory processes and inserts caudally into the ventral surface of the posterior superior iliac spine. The laterally positioned iliocostalis arises from the tip of the transverse processes and the adjacent middle layer of thoracolumbar fascia and inserts into the ventral edge of the iliac crest caudally.8 Microarchitectural studies reveal that the erector spinae muscles are designed as long muscle fascicles with relatively small PCSA. This anatomic morphology suggests that they serve to move the trunk to extension, lateral bending, and rotation. With this type of design, they are less likely to act as primary stabilizers of the vertebral column.9 Unilateral contraction of the lumbar erector spinae laterally flexes the vertebral column; bilateral contraction produces extension and posterior rotation of the vertebrae in the sagittal plane.

The interspinales, intertransversarii, and short rotator muscles are short flat muscles that lie dorsal to the intertransverse ligament (Figure 1). The intertransversarii and interspinales run along the intertransverse and the interspinous ligaments of each segment. The short rotators originate from the posterior-superior edge of the lower vertebra and attach to the lateral side of the upper vertebral lamina. Because of their small PCSA, they are not able to generate the forces needed for movement or stability of the spinal column. More likely, they act as proprioceptive sensors rather than force generating structures.1

Paraspinal Muscle Injury

Characteristics of Paraspinal Muscles in the Postsurgical Spine

Spine surgery inherently causes damage to surrounding muscles marked by atrophy and subsequent loss of function.10 Muscle atrophy coincides with decreased muscle cross sectional area (CSA), which in turn correlates with decreased force production capacity of the muscle.10–19 Because of its midline location, the multifidus muscle is most severely injured during a midline approach.20 Muscle biopsies obtained from patients undergoing revision spinal surgery exhibit an array of pathologic features that include selective type II fiber atrophy, widespread fiber type grouping (a sign of reinnervation), and “moth eaten” appearance of muscle fibers.21 Although these pathologic changes can occasionally be found in biopsies from normal individuals, the pathologic changes are more prevalent after surgery.22 Reductions in the CSA of the paraspinal muscles are greatest following a midline approach for a posterolateral fusion.17,20,23

The Mechanism of Paraspinal Muscle Injury During Surgery

The factors responsible for muscle injury during surgery have been well studied in both animals and human beings. Direct injury to the muscle is caused by dissection and stripping of tendinous attachments from the posterior elements of the spine. Additionally, extensive use of the electrocautery causes localized thermal injury and necrosis to the tissues. However, the most significant factor responsible for muscle injury is likely because of powerful self-retaining retractors. Kawaguchi et al quantified the factors responsible for muscle necrosis after a standard open midline posterior approach.24–27 They proposed that injury is caused by a crush mechanism similar to that caused by a pneumatic tourniquet during surgery of the limbs. During the application of self-retaining retractors, elevated pressures lead to decreased intramuscular perfusion.28–31 The severity of the muscle injury is correlated to the degree of the intramuscular pressure and the length of retraction time. They concluded that muscle damage can be reduced by intermittent release of the retractors during prolonged surgery combined with a relatively longer incision that allows reduced retraction pressures. In addition to crush injury, muscle function is affected by disruption of it osseotendinous attachments. A traditional open, midline laminectomy removes the spinous process. The spinous process is the sole cephalad attachment of the multifidus muscle tendon. This renders the muscle incapable of dynamically controlling its motion segment (Figure 4).

Figure 4
Figure 4:
A, An intraoperative photograph showing, minimally invasive decompression procedure for the treatment of L4–L5 spinal stenosis. The strategy is to use a bilateral paramedian approach that spares the tendinous attachment of the multifidus muscle at the spinous. B, Postoperative CT demonstrating good decompression of the spinal canal without noticeable damage to the posterior paraspinal muscles. Decompression of the left lateral recess through the right paramedian approach (and vice versa) allows minimal facet resection during decompression of the lateral recess.

Paraspinal Muscle Denervation During Posterior Lumbar Surgery

Denervation is yet another mechanism that leads to muscle degeneration and atrophy after surgery. Muscle denervation can occur in a discrete location along the supplying nerve, or be located in several points along the nerve and the neuromuscular junction. As previously described, nerve supply to the multifidus is especially vulnerable to injury because of its monosegmental innervation pattern.5 Muscle denervation is also possible through damage to the neuromuscular junction following long muscle retraction and necrosis. Shorter retraction time or an intermittent release of muscle retraction has been shown to significantly decrease degeneration and denervation of the muscles.30 Gejo et al examined the relationship between the time of retraction and postoperative damage to the paraspinal muscle, by measuring postsurgery signal intensity of the multifidus muscle, using T2-weighted magnetic resonance imaging (MRI).32 Long retraction time during surgery was found to correlate with high-signal intensity in the multifidus muscle even at 6 months postsurgery. They proposed that these findings reflect chronic denervation of the muscle caused by damage to the neuromuscular synapses. Sihvonen et al found signs of severe denervation of the multifidus muscle in patients with failed back syndrome.33 Muscle biopsies showed signs of advanced chronic denervation consisting of group atrophy, marked fibrosis, and fatty infiltration. Moreover, fiber type grouping, a histologic sign of reinnervation, was rare. They hypothesized that the denervation injury resulted from direct damage to the medial branch of the posterior rami during muscle retraction associated with the posterior midline approach.

Clinical Effects of Posterior Lumbar MIS Techniques

MIS surgery strives to minimize injury to the structures that provide dynamic stability to the spine. By eliminating the use of self-retaining retractors, intramuscular retraction pressure is reduced, leading to less muscle crush injury. Focusing the surgical corridor directly over the surgical target site allows for less muscle stripping that may otherwise disrupt its tendinous attachments or damage their neurovascular supply. Furthermore, arthrodesis strategies favor interbody fusion techniques over posterolateral fusions in a effort to avoid unnecessary paraspinal muscle stripping.

Kim et al compared trunk muscle strength between patients treated with open posterior instrumentation versus percutaneous instrumentation.34 Tests were performed isometrically at multiple flexion positions. Patients undergoing percutaneous instrumentation displayed more than 50% improvement in extension strength, whereas patients undergoing traditional midline open surgery had no significant improvement in lumbar extension strength. Extension strength correlated with preservation multifidus CSA as measured on MRI. Clinical outcomes related to hospital stay, narcotic use, and blood loss were significantly lower in the minimally invasive group. In a similar study, Stevens et al assessed the postsurgical appearance of the multifidus muscle using a high-definition MRI sequence.35 In patients treated by an open posterior transforaminal lumbar interbody fusion (TLIF) technique, marked intermuscular edema was observed on postsurgical MRI at 6 months after surgery. In contrast, patients in the MIS TLIF group had a normal appearance on MRI after surgery. The clinical results correlate well with cadaveric studies that show minimally invasive, table-mounted tubular retractors produce lower retractors pressures in the surrounding soft tissues compared with traditional self-retaining open retractors.35 By using the table mount to hold the retractor in position, the pressure with the MIS retractor was transient whereas the self-retaining retractor pressures remained elevated.35

Hyun et al retrospectively assessed a group of patients that underwent unilateral TLIF with ipsilateral instrumented posterior spinal fusion via an open technique.15 Contralateral instrumented posterior spinal fusion was performed at the same level using a paramedian, intermuscular (Wiltse) minimally invasive approach. After surgery, there was a significant decrease in the CSA of the multifidus on the side of the open approach, whereas no reduction in the multifidus CSA on the contralateral side was observed.

Decreases in tissue trauma not only have local effects but also alter overall systemic physiology. Kim et al studied circulating markers of tissue injury in patients undergoing open versus MIS fusions.36 Markers of skeletal muscle injury (creatinine kinase, aldolase), pro-inflammatory cytokines (IL-6, IL-8), and anti-inflammatory cytokines (IL-10, IL-1 receptor antagonist) were analyzed with enzyme-linked immunosorbent assay techniques. Two- to 7-fold increases in all markers were observed in the open surgery group. The greatest difference between the groups occurred on the first postoperative day. Most markers returned to baseline in 3 days for the MIS group whereas the open surgery group required 7 days. IL-6 and IL-8 are known cytokines that participate in various systemic inflammatory reactions.37,38 It is possible that such elevations in inflammatory cytokines have direct affects beyond the surgical site as persistently elevated levels of proinflammatory cytokines have been associated with organ failure in postsurgical patients.39

A recent study by Fan et al compared postoperative MRI findings, systemic creatine kinase levels, and clinical outcomes in patients undergoing open versus MIS one-level posterior lumbar interbody fusion.40 In a study of 59 patients (28 MIS vs. 31 open), the MIS group minimally invasive group had less postoperative back pain (P < 0.001) and lower postoperative Oswestry Disability Index scores (P = 0.001) at 1-year follow-up. The changes in multifidus CSA and creatinine kinase levels were positively correlated with visual analogue scale and Oswestry Disability Index scores.


The scientific basis for MIS surgery relies on a few key concepts. First, the single midline approach has been replaced by multiple paramedian approaches where the surgical corridor is placed directly over the surgical target sites in an effort to avoid detachment of the multifidus muscle tendon at the spinous process. Second, the use of self-retaining retractors, which can induce crush injuries to adjacent tissues, has been supplanted by table-mounted, tubular-type retractors that minimize pressure on muscles, vessels, and nerves. Together, these basic MIS strategies strive to minimize injury to the bones, tendons, and muscles that actively control movement and confer dynamic stability to the lumbar spine.

Key Points

  • Muscle crush injury can be avoided by using table-mounted tubular retractors that minimize retraction pressures in the adjacent soft tissues.
  • Avoid midline laminectomies that detach the multifidus muscle tendon attachment.
  • Use multiple, smaller incisions to maintain a narrow surgical corridor that uses known anatomic surgical planes.
Figure 3
Figure 3:
A, Anatomic specimen showing the spatial relationship of the medial branch of the posterior rami nerve (highlighted in black and retracted with a rubber vessel loop), the facet joint (FJ), and transverse process (TP). B, Diagram showing the path of the L1 medial branch of posterior rami nerve to the L1 multifidus muscle. The L1 medial branch of the posterior rami traverses the L2 transverse process and rises posteriorly to innervate the multifidus muscle that originates from the L1 spinous process. The L1 multifidus then separates into 4 bands which insert into the lamina and superior articular processes of the L3, L4, L5, and S1. Reprinted with permission from Spine.41 Copyright 2009, Lippincott Williams & Wilkins.


1.Bogduk N. Proceedings: the posterior lumbar muscles and nerves of the cat. J Anat 1973;116:476–7.
2.Donisch EW, Basmajian JV. Electromyography of deep back muscles in man. Am J Anat 1972;133:25–36.
3.Cholewicki J, Panjabi MM, Khachatryan A. Stabilizing function of trunk flexor-extensor muscles around a neutral spine posture. Spine (Phila Pa 1976) 1997;22:2207–12.
4.Ward SR, Kim CW, Eng CM, et al. Architectural analysis and intraoperative measurements demonstrate the unique design of the multifidus muscle for lumbar spine stability. J Bone Joint Surg Am 2009;91:176–85.
5.MacIntosh JE, Bogduk N. 1987 Volvo award in basic science. The morphology of the lumbar erector spinae. Spine (Phila Pa 1976) 1987;12:658–68.
6.Bogduk N, Macintosh JE, Pearcy MJ. A universal model of the lumbar back muscles in the upright position. Spine (Phila Pa 1976) 1992;17:897–913.
7.Macintosh JE, Bogduk N. The attachments of the lumbar erector spinae. Spine (Phila Pa 1976) 1991;16:783–92.
8.Bustami FM. A new description of the lumbar erector spinae muscle in man. J Anat 1986;144:81–91.
9.Delp SL, Suryanarayanan S, Murray WM, et al. Architecture of the rectus abdominis, quadratus lumborum, and erector spinae. J Biomech 2001;34:371–5.
10.Gejo R, Matsui H, Kawaguchi Y, et al. Serial changes in trunk muscle performance after posterior lumbar surgery. Spine (Phila Pa 1976) 1999;24:1023–8.
11.Datta G, Gnanalingham KK, Peterson D, et al. Back pain and disability after lumbar laminectomy: is there a relationship to muscle retraction? Neurosurgery 2004;54:1413–20; discussion 1420.
12.Franzini A, Ferroli P, Marras C, et al. Huge epidural hematoma after surgery for spinal cord stimulation. Acta Neurochir (Wien) 2005;147:565–7; discussion 567.
13.Granata C, Cervellati S, Ballestrazzi A, et al. Spine surgery in spinal muscular atrophy: long-term results. Neuromuscul Disord 1993;3:207–15.
14.Hutchinson D, Kozin SH, Mayer N, et al. Dynamic electromyographic evaluation of adolescents with traumatic cervical injury after biceps to triceps transfer: the role of phasic contraction. J Hand Surg Am 2008;33:1331–6.
15.Hyun SJ, Kim YB, Kim YS, et al. Postoperative changes in paraspinal muscle volume: comparison between paramedian interfascial and midline approaches for lumbar fusion. J Korean Med Sci 2007;22:646–51.
16.Mayer TG, Vanharanta H, Gatchel RJ, et al. Comparison of CT scan muscle measurements and isokinetic trunk strength in postoperative patients. Spine (Phila Pa 1976) 1989;14:33–6.
17.Motosuneya T, Asazuma T, Tsuji T, et al. Postoperative change of the cross-sectional area of back musculature after 5 surgical procedures as assessed by magnetic resonance imaging. J Spinal Disord Tech 2006;19:318–22.
18.Rantanen J, Hurme M, Falck B, et al. The lumbar multifidus muscle five years after surgery for a lumbar intervertebral disc herniation. Spine (Phila Pa 1976) 1993;18:568–74.
19.Kawaguchi Y, Matsui H, Gejo R, et al. Preventive measures of back muscle injury after posterior lumbar spine surgery in rats. Spine (Phila Pa 1976) 1998;23:2282–7; discussion 2288.
20.Gille O, Jolivet E, Dousset V, et al. Erector spinae muscle changes on magnetic resonance imaging following lumbar surgery through a posterior approach. Spine (Phila Pa 1976) 2007;32:1236–41.
21.Mattila M, Hurme M, Alaranta H, et al. The multifidus muscle in patients with lumbar disc herniation. A histochemical and morphometric analysis of intraoperative biopsies. Spine (Phila Pa 1976) 1986;11:732–8.
22.Weber BR, Grob D, Dvorak J, et al. Posterior surgical approach to the lumbar spine and its effect on the multifidus muscle. Spine (Phila Pa 1976) 1997;22:1765–72.
23.Suwa H, Hanakita J, Ohshita N, et al. Postoperative changes in paraspinal muscle thickness after various lumbar back surgery procedures. Neurol Med Chir (Tokyo) 2000;40:151–4; discussion 154–5.
24.Kawaguchi Y, Matsui H, Tsuji H. Back muscle injury after posterior lumbar spine surgery. Part 2: histologic and histochemical analyses in humans. Spine (Phila Pa 1976) 1994;19:2598–602.
25.Kawaguchi Y, Matsui H, Tsuji H. Back muscle injury after posterior lumbar spine surgery. Part 1: histologic and histochemical analyses in rats. Spine (Phila Pa 1976) 1994;19:2590–7.
26.Kawaguchi Y, Matsui H, Tsuji H. Back muscle injury after posterior lumbar spine surgery. A histologic and enzymatic analysis. Spine (Phila Pa 1976) 1996;21:941–4.
27.Kawaguchi Y, Yabuki S, Styf J, et al. Back muscle injury after posterior lumbar spine surgery. Topographic evaluation of intramuscular pressure and blood flow in the porcine back muscle during surgery. Spine (Phila Pa 1976) 1996;21:2683–8.
28.Styf J. Pressure in the erector spinae muscle during exercise. Spine (Phila Pa 1976) 1987;12:675–9.
29.Styf J, Lysell E. Chronic compartment syndrome in the erector spinae muscle. Spine (Phila Pa 1976) 1987;12:680–2.
30.Styf JR, Willen J. The effects of external compression by three different retractors on pressure in the erector spine muscles during and after posterior lumbar spine surgery in humans. Spine (Phila Pa 1976) 1998;23:354–8.
31.Taylor H, McGregor AH, Medhi-Zadeh S, et al. The impact of self-retaining retractors on the paraspinal muscles during posterior spinal surgery. Spine (Phila Pa 1976) 2002;27:2758–62.
32.Gejo R, Kawaguchi Y, Kondoh T, et al. Magnetic resonance imaging and histologic evidence of postoperative back muscle injury in rats. Spine (Phila Pa 1976) 2000;25:941–6.
33.Sihvonen T, Herno A, Paljarvi L, et al. Local denervation atrophy of paraspinal muscles in postoperative failed back syndrome. Spine (Phila Pa 1976) 1993;18:575–81.
34.Kim DY, Lee SH, Chung SK, et al. Comparison of multifidus muscle atrophy and trunk extension muscle strength: percutaneous versus open pedicle screw fixation. Spine (Phila Pa 1976) 2005;30:123–9.
35.Stevens KJ, Spenciner DB, Griffiths KL, et al. Comparison of minimally invasive and conventional open posterolateral lumbar fusion using magnetic resonance imaging and retraction pressure studies. J Spinal Disord Tech 2006;19:77–86.
36.Kim KT, Lee SH, Suk KS, et al. The quantitative analysis of tissue injury markers after mini-open lumbar fusion. Spine (Phila Pa 1976) 2006;31:712–6.
37.Igonin AA, Armstrong VW, Shipkova M, et al. Circulating cytokines as markers of systemic inflammatory response in severe community-acquired pneumonia. Clin Biochem 2004;37:204–9.
38.Baggiolini M, Dahinden CA. CC chemokines in allergic inflammation. Immunol Today 1994;15:127–33.
39.Ogawa M. Acute pancreatitis and cytokines: “second attack” by septic complication leads to organ failure. Pancreas 1998;16:312–5.
40.Fan S, Hu Z, Zhao F, et al. Multifidus muscle changes and clinical effects of one-level posterior lumbar interbody fusion: minimally invasive procedure versus conventional open approach. Eur Spine J 2010;19:316–24.
41.Regev GJ, Lee YP, Taylor WR, et al. Nerve injury to the posterior rami medial branch during the insertion of pedicle screws. Spine 2009;34:1239–42.

paraspinal muscles; retraction pressures; cytokines; dynamic stability; multifidus

© 2010 Lippincott Williams & Wilkins, Inc.