5.9 Heterotopic ossification

10.1055/b-0034-98160

5.9 Heterotopic ossification

1 Introduction to total disk arthroplasty and heterotopic ossification

The introduction of motion preserving implants in the treatment of cervical and lumbar degenerative disorders has brought forth a whole new array of observations, which may be relevant for assessment of these new technologies. Heterotopic ossification, while occasionally noted with anterior cervical plating [1, 2], had rarely been a targeted observation in spine surgery at all. The accuracy of placement of spinal hardware beyond neurologic or vascular concerns had not really been the subject of formal assessments or ratings either [3].

2 Current classification systems for heterotopic ossification

McAfee et al [4] provided the first classification system pertaining to total disc arthroplasty (TDA) specifically. The system was developed to classify the extent of heterotopic ossification (HO) following lumbar TDA and was based on the Brooker et al [5] classification for HO following total hip arthroplasty. The timing of the published classification was ideal, and was followed by the Food and Drug Administration’s (FDA) approval of the SB Charité lumbar disc prosthesis and the accompanying two part publication of the outcomes from the Investigational Device Exemption (IDE) in 2005.

McAfee’s classification of HO following lumbar TDA is reliable, simple to use, and reproducible. There are five categories described: Class 0—no HO, Class I—island of bone not within the margins of the disc and not interfering with motion, Class II—bone within the margins of the disc but not blocking motion, Class III—bone within the margins of the disc and interfering with motion of the prosthesis, and Class IV—bony ankylosis. McAfee stressed the importance of utilizing accompanying flexion-extension x-rays when utilizing the classification, and felt that previously reported rates of HO after TDA were probably inaccurate. This classification made it possible for all future reports regarding the use of total disc arthroplasty to accurately determine the incidence of HO, and will guide future research regarding the etiology of HO and its potential management.

Although McAfee’s classification of HO was described for lumbar TDA, its use has been expanded to cervical TDA. Mehren et al [6] described a modification of McAfee’s classification as it pertained to HO after cervical TDA, also utilizing flexion-extension x-rays for motion assessment, but utilizing the change in interspinous space to detect small changes in motion during the follow-up intervals. From preliminary reports of lumbar and cervical disc arthroplasty, it appears that HO is more frequent and severe in the cervical spine. It also appears to be dependent on the type of prosthesis, surgical technique, and postoperative use of nonsteroidal anti-inflammatory medications (NSAIDS).

Treatment implications

McAfee et al [7] was also instrumental in correlating the accuracy of placement of the SB Charité prosthesis and clinical outcomes in the FDA IDE prospective randomized study that led to the approval of the SB Charité in the United States. The accuracy of placement was classified into three categories: Ideal—within 3 mm of exact central placement in both the coronal and midsagittal planes (coronal plane = anteroposterior x-ray = midline or within 3 mm of midline; midsagittal plane = lateral x-ray = 2 mm posterior to middle of vertebral body or within 3 mm of this axis), Suboptimal—placement from 3 mm to 5 mm from exact central placement in at least one axis, and Poor—poor placement, defined as greater than 5 mm from exact central placement in at least one axis or any patient requiring posterior spinal fixation and arthrodesis secondary to prosthesis placement, or component subsidence or migration. The analysis revealed that accuracy of placement had a statistically significant correlation with range of motion on flexion-extension x-rays as well as clinical outcomes. Simply put, poor placement did not allow the prosthesis to biomechanically behave as intended, resulting in less preserved motion, and ultimately, poorer outcomes.

The utility of this classification, although very useful when implanting the SB Charité prosthesis, has not been shown to be universal by other investigators to date. The primary reason is that the SB Charité is biomechanically unique. It is a nonconstrained prosthesis that performs much differently than other lumbar and cervical TDA devices.

Considerations for new classification systems

A classification describing the different types of prostheses may therefore be helpful in comparing how placement accuracy of similar prostheses affects range of motion and clinical outcomes. The broad categories of nonconstrained, semiconstrained, and constrained may not be suitable for the variety of implants currently available or under investigation. In addition, the bearing surfaces may dictate ultimate performance and failure modes when imprecise placement is evaluated. This was the case with total hip arthroplasty during early stages of methylmethacrylate fixation of the prosthesis in the proximal femur. Different modes of failure were detected with different shapes and sizes of the stem, the quality of the cement mantle, and the position of the stem [8–11]. Continued follow-up with critical assessment of surgical technique, placement, outcomes, and modes of failure will help further quantify, and eventually, classify these implants into more precise categories.

Another void in the TDA literature is the classification of the preoperative diagnoses for which TDA was indicated. Similar to the range of hip pathologies that lead to total hip arthroplasty, there are differing pathologic entities which can be treated with TDA, all of which may have correspondingly different outcome and complication profiles. McAfee et al made to this in the HO classification publication. Goel et al described three different types of osteoarthritis requiring total hip arthroplasty and the risk of HO associated with each. The incidence of heterotopic bone formation was 11% in atrophic osteoarthritis, 32% in the normotrophic type, and 87% in the hypertrophic type. The difference between each type of osteoarthritis predisposing to HO of the hip was statistically significant. Although stages of disc degeneration have been classified utilizing various imaging studies, there has been no correlation with stage of degeneration and the need for surgical intervention [12–14]. Additionally, there has been no correlation made between the preoperative stage and its effect on ultimate outcomes. In the lumbar spine, the indications for lumbar total disc arthroplasty preclude the presence of significant facet arthrosis. However, there have been no reports correlating the extent or grade of facet arthrosis and outcome. Similarly, the cervical spondylitic conditions for which cervical TDA has been used is widely varied from single level soft disc herniation to complete collapse of multiple segments with loss of the physiologic lordosis. The impact of the preoperative spondylitic stage on outcomes has not been reported. There is little doubt that the preoperative status of the motion segment will influence not only the outcome, but also the eventual performance, alignment, and mode of failure of the prosthesis. Furthermore, we may find that the one type of implant performs better for a particular stage of disease.

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