Short or Long Posterior Fusion: Determining the Extent of Fixation

Luiz Roberto Vialle, Emiliano Vialle, Joana B.C.R. Guasque, and Luiz Gustavo Dal Oglio Rocha

Introduction

Thoracolumbar fractures (TLFs) are highly prevalent, and several factors influence the decision-making process regarding their treatment. Patient characteristics, the trauma mechanism, and associated injuries play a role in he decision regarding conservative versus surgical treatment. The timing of surgery, the type of approach, and determining which levels should be instrumented for achieving a successful recovery 1 are also important considerations in the decision-making process.

Traditional methods of stabilizing the injured spine entailed instrumenting two vertebrae above and two vertebrae below the injury to provide enough stabilization to enable early mobilization and return to normal activities, without incurring the risks of posttraumatic kyphosis, implant breakage, and late neurologic deficit.^1,2 A stronger fixation would also enable extensive canal decompression, including removal of fractured pedicles and vertebral body fragments, which would maximize the potential for neurologic recovery.^3,4

In the 1980s, short segment fixation (SSF) was developed using pedicle screws at the levels above and below the injury. This construct did not restore the integrity of the anterior structure, and loading the device in a cantilever mode led to a high incidence of early implant failure, loss of sagittal alignment, and nonunion.²

The evolution of surgical techniques and implants has enabled shorter fixation of TLFs, entailing a theoretical reduction of surgical time and hospital costs, in addition to sparing mobile levels at the lumbar spine.^4–6 However, with shorter fusions, a significant number of failures and late complications occurred, raising concerns about the real benefit of sparing one or two motion segments of the spine.^1,2,4 In this chapter, we discuss the rationale for short fixation for TLF, and we describe our patient evaluation method and surgical technique.

Short Segment Fixation

Short segment fixation entails including one level above and one level below the fractured segment, and sometimes including the fractured segment itself, by instrumenting the fractured pedicles or providing anterior column support.^7,8 SSF is in general a bisegmental fixation. Some injuries, such as the transosseous (Chance) fractures, can be treated with even shorter fixation by including only one motion segment.⁸

Those who advocate SSF techniques believe that less aggressive surgery avoids systemic complications, especially in already compromised polytrauma patients or in those with comorbidities.^6,9

The current technology enables surgeons to perform a SSF with small open incisions or with minimally invasive surgery (MIS) techniques when available.

Indications and Technique

Monosegmental Fixation

Monosegmental fixation entails fixing only the injured level; it provides the maximum preservation of the vertebral segments when operating on fractures. It is an ideal procedure for type A3, B1, and B2 lesions 10 that do not entail severe anterior comminution, because it cannot restore anterior column height (Fig. 7.1). A mechanical study comparing monosegmental fixation with short fixation concluded that the two models have the same stiffness.¹¹ A clinical follow-up of 60 type A3 fractures, half treated by monosegmental fixation and half by short bisegmental fixation, found that the mean operating time, the intraoperative blood loss, the postsurgical pain intensity scores on the Visual Analogue Scale (VAS), and the vertebral kyphotic angle were similar for both groups.¹²

Fig. 7.1a,b Monosegmental fixation: T11/T12 type B2 injury. (a) Preoperative radiograph. (b) Postoperative radiograph.

Bisegmental Fixation

Bisegmental fixation, or SSF, has been criticized mainly for its implant failure rate, which several researchers have tried to explain.2,13 The so-called load-sharing classification was one of the first attempts to understand which factors lead to the failure of SSF. The classification score would help surgeons decide between performing a short instrumentation or supplementing it with anterior support.¹³ Although widely used as a reference among surgeons, this score has not been validated, and some studies have challenged its application, questioning the type of implants (first-generation variable screw placement [VSP] plates, parallel-screw technique), the lack of ligamentous injury evaluation,14,15 and the absence of an adequate fracture classification system. However, the load-sharing classification score has the merit of addressing the concept of extent of vertebral body fragmentation as part of the rationale for surgical treatment. In our practice, the severity of the vertebral body fragmentation is adequately classified as a type A3 or A4 fracture.¹⁰

Some scoring systems have focused more on the need for surgery and on the approach type than on the extent of fixation, and this decision has been left mostly to the surgeon’s choice.¹⁶

It is our opinion that several factors related to surgical technique lead to a successful short fixation (Fig. 7.2): adequate injury classification; proper patient positioning; and the use of the ligamentotaxis technique, divergent screws, specifically designed implants, and load sharing at the fractured vertebra.

Fig. 7.2a,b (a) L2 fracture, type A3. (b) Short segment fixation (SSF) with the internal fixator and Schanz pins.

Adequate Injury Classification

Over the years, several classifications have tried to provide a clear fracture interpretation from images and clinical data. A correct injury classification is crucial to understanding the injury’s attributes, to determining the appropriateness of surgical or nonsurgical treatment, and to determining what is specifically demanded for reduction and stabilization. This book follows the new AOSpine Thoracolumbar Fracture Classification,10 which helps to define the ideal fracture types for short- or long-segment fixation. Spinal injuries associated with high translational forces may benefit from long-segment fixation. These injuries are type C injuries as well as type B injuries associated with high vertebral body comminution; these injuries cannot be corrected by ligamentotaxis and require strong anchoring for reduction and stabilization.

Patient Positioning

Patient positioning is a crucial step in fracture reduction, and we recommend placing patients on a four-post frame, which enables an initial postural reduction by the lordotic posture and applies tension to the anterior longitudinal ligament. Patients who are placed in a kyphotic position, over towel rolls, for example, are less likely to achieve a good reduction, leading to a higher risk of fixation failure. This postural or lordotic reduction applies mainly to type A fractures and some type B fractures, but it might be dangerous for type B3 fractures.

Ligamentotaxis

This is an orthopedic procedure in which fracture fragments are brought together through the tension of ligaments and the periosteum attached to them. Such tensioning maneuvers are crucial for the success of SSF, and their final results help the surgeon to determine whether successful reduction was achieved or if other techniques should be employed.

Fractures with a ligamentous injury are not amenable to ligamentotaxis; this includes type C and some type B injuries, as well as an uncommon type A4 fracture in which the vertebral body fragment rotates 180 degrees, indicating rupture of the posterior longitudinal ligament. This injury can be identified by what is known as the reversed cortical sign. Some of these type A4 injuries can still be treated by SSF if good decompression and reduction are achieved.

Divergent Screws and Implants

The insertion of parallel pedicle screws, such as those used for deformity of degenerative disorders, places most of the load over the pedicle, which increases the risk of implant breakage if fracture healing does not occur as expected or if reduction was not satisfactory. Furthermore, polyaxial pedicle screws limit the amount of correction by cantilever maneuvers due to their natural adjustment to the rod; some loss of correction by movement of the screw–head interface can also be expected.¹⁷

The use of a divergent bridging construct, rather than parallel tension band construct, facilitates the bone–screw interface and enables a decreased compressive force at the anterior column. But what needs to be highlighted with this technique is that it reduces the mechanical stress at the rod–screw junction, in comparison with the right-angle attack between a standard screw–rod implant. This is what may prevent the implant breakage. Ouellet et al¹⁸ have demonstrated the biomechanical superiority of such a construct in their study. They have also determined the best trajectory for screw positioning; the screw trajectory is planned as a line from the anterosuperior/anteroinferior corner of the vertebral body running 5 mm above/below the pedicle cortex, allowing at least for a 6-mm screw (Fig. 7.3).

The fundamental principle to this technique—an oblique angle between the screw and the rod—makes its application with conventional pedicle screws difficult. The rods have to be contoured in lordosis, and this sometimes might not match the thoracolumbar anatomy. We have used a system with an independent screw–rod connector (Internal Fixation Systems), which enables several degrees of freedom in the screw–rod junction during the correction maneuvers. This internal fixation system and other similar devices designed specifically for TLFs have the ability to mobilize the spine with independent correction forces—lordosis and distraction (Fig. 7.4)—correcting both compressive (type A) and distraction (type B) fractures in a very stable way. These devices can also indirectly decompress the spinal canal through ligamentotaxis. We strongly recommend the selection of these implants when considering the treatment of type A and type B TLFs with a short fixation technique (Fig. 7.5).

Fig. 7.3 Insertion of divergent Schanz pins.

Before divergent screws were used, correction was dependent on positioning and in-situ rod bending, which placed high stresses on the bone–screw interface, sometimes requiring extension of the fixation with more screws. There are reports on the use of hooks to protect the screws, but they entailed a relatively high failure rate.19,20

Load Sharing at the Fractured Vertebra

Some fractures are extremely comminuted, or not completely reduced, challenging the implants and leading to a significant risk of fixation failure. Adding instrumentation to the fractured segment, either by placing pedicle screws at the fractured level or by anterior structural support, would reduce the risk of failure and maintain the benefits of a short fixation.^21,22 This technique may increase stability in 31% according to the biomechanical cadaveric study performed by Norton et al.²³ Clinical reports have shown minimal loss of correction in comparison to short fixation without intermediary screws.^21,24,25 Because the instrumentation of both pedicles could render an eventual anterior decompression impossible, we have used a single screw on the right pedicle with results similar to those presented in the literature^26,27 (Fig. 7.6).