Rigid versus Dynamic Cervical Plates: Indications and Efficacy

PAUL E. KRAEMER AND RICK C. SASSO

Over the course of 50 years of anterior cervical diskectomy and fusion (ACDF),1 graft-related complications^2,3 have remained among the most difficult problems plaguing an otherwise highly successful operation.⁴ Graft settling, fractures, dislodgment, nonunions, graft harvest site complications, cervical kyphosis, and other similar problems^2,3 predate the development of cervical instrumentation, and have been major forces driving the evolution of cervical plates. Graft choice continues to be controversial,⁵ though a general trend away from iliac crest autograft (ICBG) and toward various commercially available allograft products has permeated spinal surgery. Recent studies have compared various graft choices,⁵ and a full discussion is beyond the scope of this chapter.

The use of cervical plates has become well accepted, with multiple studies showing an increase in fusion rates, a decrease is graft dislodgment, faster speed of fusion, and less need for external postoperative immobilization.^6,7 As with many innovations, there have been unintended consequences.^8–10 A full discussion of plate-related complications is also beyond the scope of this chapter.

Cervical plates have evolved significantly from the earliest designs,⁹ which were very similar to basic rigid plates used in long bone fixation. There was no mechanism for preventing backout of the unicortical screws, which tended to be the predominant failure mode.¹⁰ Bicortical fixation helped obviate some of those concerns but had its own attendant issues. Several mechanisms have subsequently been developed to prevent the screw from backing out by mechanically blocking the head or by rigidly attaching the screw to the plate in a fixed angle manner. Both of these design concepts have been successful in preventing backout. Another design concept related to screw capture was the angle at which the screws must be inserted in relation to the plate. Some designs have a fixed angle at which the screw must be inserted, whereas others allow some variability while still capturing the screw head. Some of these plates in the second category allow for continued settling as the graft loses height, which allows continued load sharing through the graft rather than shielding the forces by the plate.^11,12

Two specific design concepts must be defined and compared with static plates. A dynamic plate may be either translationally (axially) dynamic, angularly dynamic, or both. Plate designs can be angularly dynamic if they have no constraining mechanism to control the angle of the screw in relation to the plate, as described earlier. This allows a small amount of toggle as the interbody graft collapses and allows continued compressive forces to be seen by the graft. These plates generally incorporate this angular dynamization while incorporating design characteristics that prevent screw backout. Translationally dynamic plates have a design that allows controlled cephalocaudal translational collapse either through slotted holes or a collapsible plate. A plate may potentially incorporate both translational and angular design characteristics¹³ (Fig. 22.1).

With multiple designs over multiple generations and multiple confounding factors, there are key questions left unanswered, which may include the optimum graft choice in each unique clinical situation, the relationship of fusion to clinical efficacy, the relative advantage of preservation of local lordosis weighed against fusion rates, and plate-specific design for multilevel constructs. This review explores data specifically regarding fusion rate and clinical success, with mention of plate-specific complications.

Description of Search

Searching Ovid and PubMed for the terms “dynamic” and “cervical” and “plate” returned 74 results. Multiple studies were immediately rejected because the term “dynamic” referred to flexion-extension films rather than plate design. The majority of remaining studies specifically referencing dynamic plates were either reviews or biomechanical studies with no specific clinical investigation. Clinical studies were each investigated, and references from these studies backtracked for a full picture of the literature available on dynamic cervical plates. An additional search on “ACDF and dynamic,” and “cervical fusion and dynamic” yielded no further results.

Fig. 22.1 (A) Atlantis Vision System cervical plate, which incorporates angular dynamization, and (B) Atlantis Translational cervical plate, which incorporates angular and translational dynamization. Images provided by Medtronic Sofamor Danek USA, Inc. Atlantis Vision System, and Atlantis Translational System incorporate technology developed by Gary K. Michelson, MD.

A single-level I study was identified. A single-level II study was identified. Six level III and IV studies were reviewed.

Level I Data

The only well-designed multicenter, randomized, clinical trial with independent analysis has been published recently (April 2009) out of four centers in central Europe.14 Pitzen and colleagues present 2-year follow-up on a prospective investigation comparing one-or two-level ACDFs utilizing static versus translationally dynamic plates. The authors report on plate complications, fusion rates, speed of fusion, lordosis, and clinical outcomes, including neck disability index (NDI) and visual analogue scale (VAS). They randomized patients prior to surgery, used identical technique including iliac crest autograft, and utilized two well-established plating systems, one static (CSLP, Synthes, Paoli, PA) and one dynamic (ABC Plate, Aesculap, San Francisco, CA). Clinical results at 2 years showed no differences in fusion rate using any of the methods employed to determine bony fusion, although segmental mobility on flexion-extension films was significantly less (indicating fusion) at 6 months in the dynamic group. Loss of lordosis was somewhat increased in the dynamic group, at 4.3 degrees versus 0.7. Implant failure was present in 4/63 static plates, whereas none of the dynamic plates suffered a complication. Importantly, no clinical measures (VAS, NDI, narcotic usage) revealed any differences. There was no statistical difference noted between single-versus two-level procedures in either group. No graft-related judgments may be drawn because all procedures utilized iliac crest autograft (ICBG).

Level II Data

A single-center blinded, randomized trial from 2007 reports on the results of static versus translationally dynamic (slotted hole) plates. Nunley et al¹⁵ prospectively investigated 66 patients undergoing one-to three-level ACDF with allograft cortical bone as the graft choice. Average 16-month (minimum 12) NDI, VAS, and radiographic assessment of fusion were the primary outcomes measures, and clinical and radiographic results were correlated. The authors reported no significant overall differences in clinical outcomes with either VAS (p = 0.49) or NDI (p = 0.31); however, they did note a subgroup analysis showing multilevel fusions with a dynamic plate having better outcomes (p = 0.05). Fusion rates by plate and number of levels were not separated out. The study is not awarded a level-one rating despite being a prospective randomized, controlled trial due to inadequate sample size, no independent examination of radiographic results, and imbalance within the subgroups for which significant results were reported. However, longer constructs may benefit from a dynamized plate, and this trial continues to accept enrollees.

Level III Data

A level III prospective cohort study from Croatia¹⁶ has recently compared three separate constructs in patients undergoing ACDF at one or more levels. Over half the patients were a single level, but subjects undergoing up to four-level fusions without posterior augmentation were included. A translationally dynamic plate and an angularly dynamized plate were compared, along with nonplated fusions. Seven of 33 patients with the dynamic plate had superjacent heterotopic ossification (a described complication of some designs, likely from the sliding mechanism of the plate impinging on the level above and stimulating bone formation¹³); however, this plate was associated with the fastest fusion. Two of 33 angularly dynamized plates failed to fuse, for a 94% fusion rate. Three of 15 in the nonplated group had graft extrusion problems. This is the only paper directly comparing the two classes of plates. The authors drew the conclusion that plate choice did not have a significant impact on clinical outcome.

Table 22.1 Summary of Published Studies

*Study*	*Level*	*Description*
Pitzen et al¹⁴	I	RCT showing no clinical or radiographic differences but lower failure rate in dynamic plate
Nunley et al¹⁵	II	RCT with low numbers, multilevel subgroup was better with dynamic plate
Stancić et al¹⁶	III	Prospective series comparing two dynamized designs to nonplated subjects; no clinical differences
DuBois et al¹⁷	III	Small nonrandomized comparison of multilevel ACDF, better fusion rate with static plate over angular dynamized
Kim et al¹⁸	III	Retrospective series showing better fusion rate with dynamic plate
Epstein¹⁹	IV	Case series of multilevel corpectomies, low complication rate
Epstein²⁰	IV	Case series of OPLL corpectomies
Epstein²¹	IV	Case series of single-level ACDFs

Abbreviations: ACDF, anterior cervical diskectomy and fusion; OPLL, ossified posterior longitudinal ligament; RCT, randomized, controlled trial.

Dubois et al 17 performed a retrospective review of multilevel ACDFs (two- and three-level procedures, no corpectomies) over a 4-year period, comparing a statically locked plate (Orion plate, Medtronic Sofamor Danek, Memphis, TN) versus an angularly dynamized plate (Atlantis plate, Medtronic Sofamor Danek, Memphis, TN). Graft choice was split between autograft and structural allograft. Though patients were not randomized, no significant differences were found between the groups. Clinical results at final follow-up were similar between the groups, though the pseudarthrosis rate was higher in the dynamic group (16% vs 5%). Graft choice and smoking status were not related to clinical or radiographic success.

Kim et al¹⁸ presented a similar retrospective review of the Orion plate (Medtronic Sofamor Danek) versus the ABC plate (Aesculap, San Francisco, CA) in predominantly single level disease. They reported a higher fusion rate (97% vs 90%) and no plate-related complications in the dynamic plating system (vs 3/31 in the Orion). Again, there were no differences in clinical results.

Epstein has evaluated the clinical results of a translationally dynamic plate (ABC plate, Aesculap) in multilevel corpectomies for severe myelo/radiculopathy,¹⁹ showing significant improvement as assessed by Nurick grade, Odom’s criteria, and Short Form-36 (SF-36) scores. There was no comparison group. There was a single (2.5%) pseudarthrosis, which compares favorably with the literature. A subgroup of the same patients from the same author²⁰ evaluated the failure rate of dynamic versus static plates in multilevel corpectomy procedures for ossified posterior longitudinal ligament (OPLL), finding a significant decrease utilizing a dynamic plate (3.6%) versus static (13%). Important considerations in both studies include the technique (multilevel anterior corpectomy, posterior wire fixation, and halo immobilization for > 4 months), and in the second study mentioned, the disease process (multilevel OPLL), which may not be generalizeable to many surgeons’ practice.

This author also reported on a small consecutive series²¹ of single-level dynamic plating. Complications were seen in four patients (9.5%), including two pseudarthroses, one plate failure, and one graft fracture. There was no comparison group.

Summary Statement

There is a paucity of evidence on relative advantages of static versus dynamic plates. A single recent level-one study is the best available evidence and seems to be concordant with the trend of lower quality of evidence investigations, which suggest that dynamic plates may provide excellent results comparable with static plates and may achieve a small increase in fusion rate, though this has not been fully consistent. Plate-specific failure rates, loss of lordosis, and failure mechanisms remain to be fully elucidated but do not appear to be markedly problematic. Angular versus translationally dynamic plate choice has not been fully explored, though the only study in which fusion rate was lower than a static plate was with an angular system. Longer constructs may benefit from dynamic plates, though, as with the remainder of this topic, would benefit from further well-designed studies. It is important to note that no study has demonstrated a significant difference in clinical results based on plate choice (Table 22.1).

Pearls

• Both static and dynamic cervical plates can be associated with high fusion rates and a low complication rate when used to augment fusion after anterior cervical diskectomy or corpectomy.

• Dynamic cervical plates may be associated with a higher rate of fusion and a lower rate of plate-specific complications; however, there is no evidence that dynamic plates offer a clinical advantage over static plates.

Only gold members can continue reading. Log In or Register to continue