History of Spinopelvic Fixation

The development of spinopelvic fixation arose soon after the introduction of Harrington rods for the treatment of scoliosis.^7–9 Harrington rod instrumentation was simple, consisting initially of hook-rod distraction. Two major problems became evident when attempting to fuse the lumbosacral spine with this instrumentation. First was the high rate of pseudarthrosis and second was the rate of sacral hook dislodgment.¹⁰ In response to these problems, Luque introduced the “L” rod, with the distal portion of the rod shaped to penetrate the iliac crest. Ferguson and Allen further modified the Luque system of segmental instrumentation by inserting angled rods into the iliac bones and passing them into hard cortical bone above the sciatic notch to achieve a rigid fixation of the lumbosacro-pelvic spine.^11–14 This fixation method became known as the Galveston technique and increased biomechanical strength by extending use of instruments into the pelvis (Fig. 296-1).^15,16 This technique became an ideal adjunct for long segment fusions in scoliosis and pelvic obliquity. However, the inherent limitations, primarily related to complex three-dimensional contouring of the rod before its insertion into the ilium, prompted the development of other techniques for lumbosacropelvic fixation. The technical challenge of contouring rods for the Luque-Galveston technique was overcome by the introduction of iliac screw instrumentation.^17–19 These fully threaded iliac bolts or polyaxial iliac screws may be connected to longitudinal rod constructs either directly or with offset connectors. They have the additional advantage of modularity, facility of placement, and the capacity to place up to two screws on each side. Furthermore, the biomechanical properties of a threaded screw design, as opposed to the smooth Galveston rod, makes it less prone to pull out.

FIGURE 296-1 Illustration of the Luque-Galveston rod technique demonstrating the distal portion of the rod contoured at an angle into the ilium. Although this technique reduced the rate of pseudoarthrosis seen in Harrington rod instrumentation, it introduced the added complexity of three-dimensional contouring. The introduction of the iliac screw precluded the need for such contouring.

Anatomy of the Lumbosacral Spine and Pelvis

The anatomy of the lumbosacral spine and pelvis is an intricate and multifaceted osseous structure, bound by multiple ligaments and tendinous insertions externally while intertwined with viscera, vessels, and neural elements from within. The sacrum unites the two hemipelvises and serves as the base for the lumbar spine. The critical anatomy that is relevant for instrumentation of the lumbosacral region is also centered at the sacrum. The internal iliac artery and vein, middle sacral artery and vein, sympathetic chain, lumbosacral trunk, and rectum all lie directly over this structure and remain vulnerable to injury during instrumentation of this region (Fig. 296-2).²

FIGURE 296-2 Anatomy of the pelvis. This figure demonstrates the viscera, vessels, and neural elements within the pelvic bowl, in particular the internal iliac artery and vein, middle sacral artery and vein, sympathetic chain, lumbosacral trunk, and rectum.

The pelvis in its entirety is composed of the sacrum, ischium, pubic bone, and ilium. The ilium is demarcated posteriorly by the posterior superior iliac spine, which is an important landmark for iliac instrumentation (see Operative Technique section). The most rostral point of the iliac crest represents the thickest cancellous bone, which tapers incrementally in the caudal direction where there is cortical bone bordering the greater sciatic notch. The understanding of this anatomy is important to achieve solid purchase for iliac screws.²⁰

The pelvic unit consists of two hemipelvises (connected anteriorly at the pubis) and the sacrum (connected posteriorly via the sacroiliac joints). The sacroiliac joint functions primarily to transfer axial loads to each hemipelvis and allows almost no motion.²¹

The bony architecture of the sacrum is distinct from that of the lumbar spine. The most rostral aspect of the sacrum has an anterior-posterior diameter of 45 to 50 mm, which tapers to 20 to 30 mm at its most caudal point.²⁰ With the exception of the cortical bone found in the sacral promontory and alae, the sacrum has a primarily cancellous architecture with pedicles that are broader and less favorable to pedicle fixation than in the lumbar spine.^3,18

Biomechanics of the Lumbosacral Spine

Transition from the mobile lumbar spine to the rigid sacropelvis results in large flexion forces at the lumbosacral pivot point. These flexion forces stress lumbosacral fixation and can lead to screw pull-out and/or pseudarthrosis.¹⁶ Augmentation of lumbosacral constructs with the addition of iliac screws may decrease the strain on sacral pedicle screws by up to 300%.

The distinct anatomic and biomechanical properties of the sacropelvic region allow for its conceptual division into three zones: zone 1 includes the proximal sacrum, zone 2 includes the alar wings to the distal sacrum, and zone 3 includes the ilium. Instrumentation of zone 1 typically involves S1 pedicle fixation. Given the broad cancellous pedicles unique to the sacrum, the trajectory of these screws should converge toward the tricortical point of the sacral promontory for optimal fixation. In zone 2, alar screws may be used to augment zone 1 instrumentation and thereby increase sacral fixation by up to 20%. Finally, the most significant biomechanical anchor for sacral fixation is provided by instrumentation of zone 3, the ilium. Iliac screws provide strong resistance to flexion and pull-out forces and are ideal to augment L5 and S1 screw fixation (Fig. 296-3).²²

FIGURE 296-3 Zones of the pelvis. The distinct biomechanical properties of the sacropelvic region allow for its division into three zones: zone 1 includes the proximal sacrum, zone 2 includes the alar wings to the distal sacrum, and zone 3 includes the ilium.

(Adapted from Kuklo TR, O’Brien MF. Sacropelvic fixation options. In: Mummaneni PV, Haid RW, Lenke LG, eds. Spinal Deformity: A Guide to Surgical Treatment and Planning. St. Louis: QMP, 2008:647-664.)

Operative Technique for Pelvic Screw Fixation

The patient is positioned prone on bolsters that support the chest and anterior pelvis (Table 296-1). Care is taken to position the lumbar spine in a lordotic position to minimize “flat back” syndrome. A midline incision is made above the level to be fused and extended downward to the bottom of S2. Subperiosteal dissection of the lumbar paraspinous muscles is performed with Bovie cautery and periosteal elevators. We prefer to place the hardware cephalad and do any necessary decompression of the neural elements before exposing the midsacrum and posterior superior iliac spine to limit blood loss from muscle exposure. Subsequently, subperiosteal dissection is extended over the midsacrum and laterally to expose the medial overhang of the ilium. We take care to minimize cautery around the dorsal sacral foramina to prevent bleeding and we do not cauterize into the sacroiliac joint.

TABLE 296-1 Operative Technique for Sacropelvic Fixation

1 Expose the posterior superior iliac spine. 2 Locate the entry point of the pelvic screw (iliac screw). a 1 cm deep to the superficial edge of the posterior superior iliac spine b 1 cm proximal to the palpated inferior end of the posterior superior iliac spine 3 Decorticate the entry point with a drill or awl. 4 Use a gear shift probe to create the pilot hole for the screw. a Aim for the thick bone just above the greater sciatic notch i Typically 30-45 degrees down angle in the axial plane and 30-45 degrees inferior in the coronal plane ii Typical depth, 60-80 mm iii May use anteroposterior, pelvic inlet, and obturator fluoroscopic views to guide trajectory iv May expose superficial surface of ilium to guide screw trajectory in the axial plane 5 Tap pilot hole (undersize tap). 6 Probe for cortical wall violation. 7 Place pelvic screw. a Recess head into posterior superior iliac spine 8 Connect to lumbosacral rod inferior to S1 level. a May use connector

The posterior superior iliac spine, including the distal overhang of this structure over the sacrum, is exposed. The entry point for the pelvic screw, which is 1 cm rostral to the palpated inferior overhang of the posterior superior iliac spine and 1 cm below the superficial ridge of the posterior superior iliac spine, is identified (Fig. 296-4, A). It is important to place the screw head deep to the superficial ridge of the posterior superior iliac spine because this structure is the most prominent bony structure that individuals feel when they sit against a hard backed chair (Fig. 296-4, B).

FIGURE 296-4 Operative technique for spinopelvic fixation. A, Illustration of a pelvic gear shift probe inserted at the entry point for pelvic screw fixation: 1 cm rostral to the inferior end of the posterior superior iliac spine and 1 cm deep to the superficial ridge of the posterior superior iliac spine. B, Illustration of completed spinopelvic fixation construct. Note that the polyaxial pelvic screw heads are buried into the posterior superior iliac spine deep to its superficial ridge to prevent hardware prominence.

Following the exposure, the pelvic screw entry point is decorticated with a small drill. The pelvic gear shift is then placed into the entry point and aimed toward the thick bone, which is just superior to the greater sciatic notch and potentially onward toward the anterior inferior iliac spine, taking care not to violate the top of the acetabulum. This optimizes the bone to screw interface. The gearshift is gradually advanced 60 mm or more to reach the target bone above the greater sciatic notch. Care must be taken not to enter the greater sciatic notch itself to avoid injuring the neurovascular structures that traverse it (i.e., the superior gluteal vessels and sciatic nerve). Anteroposterior fluoroscopy typically provides adequate guidance to direct the pelvic gear shift to the bone just superior to the greater sciatic notch. The angle of the gear shift from the entry point is typically 30 to 45 degrees inferiorly in the coronal plane and 30 to 45 degrees anteriorly in the axial plane (see Fig. 296-4, A).

A common problem encountered during advancement of the pelvic gear shift is to violate the superficial cortex of the ilium before reaching the target bone above the greater sciatic notch. This typically occurs because the pelvic gear shift is directed too superficially in the axial plane. Another potential pitfall is entry of part of the probe into the sacroiliac joint with subsequent violation of the anterior cortical wall of the pelvis. Early in our experience, we dissected the gluteal musculature off the dorsal surface of the ilium lateral to the posterior superior iliac spine in order to palpate the “downslope” of the ilium to guide our axial angle. However, as our experience increased, we ceased this added exposure. Fluoroscopy can be of assistance in lining up the pelvic screw. Specifically, pelvic inlet views and obturator views may help to guide the pelvic screw down the intended trajectory.

After the pilot hole in the ilium is created with a gear shift, we then probe the pilot hole with a ball-tipped feeler probe to ensure that we have no cortical violations in the screw path. We then tap the hole with an undersized tap (typically 6 mm in width) and then place the pelvic screw (typically 7 to 8 mm in width and 65 to 80 mm or more in length). We take care to avoid violating the sacroiliac joint with this and subsequent steps.

Subsequently, the pelvic screw is then threaded into the pilot hole until the head of the screw is recessed into the posterior superior iliac spine. This step in screw placement is important because recession of the screw head into the bone prevents the patient from feeling a prominent screw head when he or she sits against a hard-backed chair. It is important to note that some prominence of the pelvic screw heads may be unavoidable in thin patients who have limited muscular covering of the lumbosacropelvic area. However, burying the screw heads below the superficial portion of the posterior superior iliac spine helps to prevent superficial skin breakdown and pressure sores in these patients.²³

The final step is to connect the pelvic screw to the lumbosacral rod. It is important to create the pelvic screw connection to the lumbosacral rod inferior to S1 to avoid interference of rod linkage with the S1 screw (if present). In the past we have used iliac bolts (fixed angle iliac screws) to fixate the pelvis and connect them with closed connectors to the lumbosacral rod. However, we now prefer to use polyaxial pelvic screws, which can easily link to the rod either directly or via an open top connector or closed connector (Expedium, Depuy Spine, Raynham, MA).

Indications for Spinopelvic Fixation

See Table 296-2.

TABLE 296-2 Indications for Extension of Lumbosacral Fusions to the Ilium

1 Grade 2 or higher L5-S1 spondylolisthesis 2 Long segment fusions to the sacrum (L2 or above to S1) a Spinal deformity (scoliosis or kyphosis) b Lumbar fractures i Trauma ii Osteomyelitis iii Neoplasm 3 Lesions that destroy the sacrum a Neoplasm b Osteomyelitis c Fractures 4 Treatment of L5-S1 pseudoarthrosis

Long Segment Fusion for Scoliosis

Traditionally, long segment fusions in patients with idiopathic scoliosis with extension to the sacrum have resulted in high rates of pseudarthrosis at the S1 level.^14,24,25 Similar patterns have been identified in adult degenerative lumbar scoliosis.^26,27 As mentioned above, long segment fusions to the sacrum place a significant strain on the S1 pedicle screws, and these screws can loosen or fail with flexion maneuvers.^2,6,28,29 The addition of iliac screws significantly reduces the strain exerted upon the S1 screws.⁶ In a retrospective review of scoliosis fusions, Islam and coworkers reported a pseudarthrosis rate of 53% in those patients in whom fusion was limited to sacral fixation compared with 21% when both sacral and iliac fixation points were used.²⁵ In another analysis of long segment fusions and high-grade spondylolisthesis, Kuklo and associates reported an overall fusion rate of 95.1% with bilateral sacral screws augmented with bilateral iliac screws.¹⁹

CASE 1

Scoliosis

A 72-year-old woman previously was treated with lumbar laminectomies of L4 and L5 for lumbar stenosis and subsequently underwent placement of a spinal cord stimulator. The patient reported back pain and difficulty standing erect. Thus the patient underwent a long segment correction and fusion of her kyphoscoliosis (Fig. 296-5, A and B). Given the large flexion forces that would result from the length of this construct, the decision was made to extend the instrumentation to the pelvis. The patient underwent T3-S1 pedicle screws with bilateral pelvic screws followed by L4-5 and L5-S1 anterior interbody fusion.

FIGURE 296-5 Indications for spinopelvic fixation: kyphoscoliosis. To correct the kyphoscoliosis (A) and (B), the fusion construct was extended to the pelvis and further augmented with anterior interbody constructs (C) and (D).

High-Grade Spondylolisthesis or Pseudoarthrosis Following Fixation for L5-S1 Spondylolisthesis

Fusion rates in high-grade spondylolisthesis patients at the L5-S1 level are lower than fusion rates for degenerative spondylosis in the lumbar spine.^30,31 The increased biomechanical forces, widened sacral pedicles, and limited capacity for interbody cage placement are believed to be among the contributing factors in the elevated rate of pseudarthrosis in L5-S1 high-grade spondylolisthesis cases.³²

Several authors have previously shown the efficacy of augmenting a lumbosacral pedicle screw construct in high-grade listhesis cases with iliac fixation.^{4,17,19,30,31} Bridwell and coworkers reported their experience in salvaging pseudarthrosis cases with L5-S1 spondylolisthesis and concluded that L5-S1 pedicle fixation augmented with anterior interbody fusion and iliac fixation was a successful treatment strategy for these challenging cases.

CASE 2

High-Grade Spondylolisthesis

A 37-year-old patient with lumbar grade 2 to 3 spondylolisthesis had previously undergone L4 and L5 laminectomies. The patient then presented to us with worsening bilateral lower extremity symptoms with radiographic progression of listhesis. High flexion forces are present at the sacrum in cases of high-grade spondylolisthesis; hence, the operative management strategy for this patient included pedicle screw fixation from L3 to S1 with extension to the pelvis. This construct was further augmented by placement of an L5-S1 cage (Fig. 296-6).

FIGURE 296-6 Indication for spinopelvic fixation: high-grade spondylolisthesis. Sagittal T2 MR of lumbar spine. The grade III spondylolisthesis (A) creates increased flexion forces at the lumbosacral junction and limits the amount of interbody contact. Extension of the construct to the pelvis with iliac screws decreases the strain on the sacral pedicle screws.

Sacral Osteolytic Neoplasms and Osteomyelitis

Osteolytic destruction of the lumbar and sacral bony elements in patients with a neoplastic or infectious process presents a unique challenge for stabilization of the lumbosacral spine.

When a neoplastic or infectious process has eroded the pedicle fixation points at S1, stabilization must be considered beyond the sacrum. Lumbopelvic fixation may be the only viable means of stabilization. Gokaslan and others have reported their experience in the stabilization of similar scenarios, primarily using a modified Galveston L-rod technique.^5,33,34 We have successfully used iliac screw fixation in such cases.³¹

It is important to note that in many cases, iliac screw fixation may loosen over time because fusions are not performed across the length of the sacroiliac joint. Often, the purpose of iliac screws is to stress-shield S1 pedicle screws until fusion has occurred across the lumbosacral junction. Only in cases where S1 screw fixation is not possible (due to tumor or infection) do we make an attempt to fuse from the lumbar spine onto the medial ilium (which requires a significant volume of bone graft).

CASE 3

Osteomyelitis

This 53-year-old patient (Fig. 296-7) had advanced osteomyelitis of the L5-S1 disk space, with near-complete destruction of the L5 and S1 vertebral bodies and pedicles. The absence of fixation points at L5 and S1 prompted the use of instruments in the ala and ilium after débridement of the phlegmon.

FIGURE 296-7 Indications for spinopelvic fixation: destructive osteomyelitis. When a neoplastic or infectious process has eroded the pedicle fixation points at S1, stabilization must be considered beyond the sacrum. Lumbopelvic fixation may be the only viable means of stabilization under these circumstances. A, Sagittal MRI T1 with contrast and (B) sagittal CT reconstruction of lumbar spine demonstrating near complete obliteration of the L5 and S1 pedicles because of advanced osteomyelitis. C and D, anteroposterior and lateral views of the construct extending from the pedicles of L3 and L4 to pelvic instrumentation into the ala and ilium.

Complications

As with any instrumented fusion, hardware failure, pseudarthrosis, and screw pull-out are observed complications when instrumenting the sacropelvis. The risk of pseudarthrosis at S1 may be minimized by aiming the S1 screws at the sacral promontory to achieve anterior cortical fixation. In addition, anterior column augmentation with interbody cage support is also helpful to minimize pseudarthrosis.

Given the number of vascular, neural, and enteric structures contained within the sacropelvic bowl, misplaced iliac screws may have substantial morbidity. For instance, errant iliac screws could potentially injure structures within the greater sciatic notch or violate the acetabulum. Nevertheless, vascular or neural injuries have not been reported among the large series of pelvic screw fixations in the literature.

The immediate perioperative complications are related to the extensive surgical exposure that may be required for this technique. For example, risk of infection (approximately 4%) and high volumes of blood loss have been reported with long segment spinal fixation extending to the pelvis. Midterm complications have been related to prominence of the hardware, which can be avoided by recessing the screw heads into the iliac spine, ensuring adequate soft tissue coverage.

Conclusions

Sacropelvic instrumentation augments the rigidity of a sacral construct in situations of increased biomechanical stress at the sacral fixation site. Moreover, iliac fixation allows for caudal fixation in cases where the sacrum is destroyed by neoplasms or osteomyelitis. Although sacropelvic fixation is not indicated as part of routine lumbosacral fusions, it enhances the rigidity of long segment fusions (L2-S1 or longer constructs) for correction of scoliosis and further stabilizes the lumbosacral junction in pseudarthrosis and high-grade spondylolisthesis cases.