Cervical deformity and malalignment is a topic that has been receiving increased interest in the recent years. It has become apparent that a significant number of patients with thoracolumbar adult spinal deformity have concomitant cervical deformity. 1,2,3,4 As more attention has been paid to compensatory changes in spine surgery, there has been a growing understanding of reciprocal changes in cervical alignment with thoracolumbar deformity surgery. 5,6 There has been a growing appreciation of the fact that these changes in alignment can have a real correlation with patient disability.
The importance of cervical alignment is not surprising given the paramount importance of horizontal gaze in day-to-day life. Many of the compensatory mechanisms described in the adult deformity literature serve one of two purposes: (1) to enable standing within a cone of economy and (2) to maintain horizontal gaze. 1,7,8 While the thoracolumbar spine responds to the first goal, the cervical spine must often make reciprocal changes to enable the second. 8 These reciprocal changes, however, can sometimes result in neck pain, dysphagia, or other complaints that can cause significant disability. This chapter seeks to organize the existing literature on cervical malalignment and its impact on disability as quantified by health-related quality-of-life (HRQOL) scores.
6.2 The Challenge of Measuring Disability in the Cervical Spine
While there is no question that cervical deformity can cause significant disability, measuring the degree of disability can frequently be challenging. This difficulty represents a key limitation of much of the literature discussed in this chapter.
Currently, the disability caused by cervical pathology is measured using a number of disease-specific patient-reported outcome (PROs) measures or surgeon-administered tools to evaluate functional status. 9,10 None of these measures, however, have proven themselves to be a “gold standard” for cervical pathology. 11,12 Examples of widely used PROs include the Neck Disability Index (NDI), the neck and arm pain Visual Analog Scale (VAS), and the 36-Item Short-Form Health Survey (SF-36). Examples of investigator-administered tools include the modified Japanese Orthopedic Association (mJOA) and Nurick scales.
The NDI, the most commonly used PRO, 13 illustrates many of the limitations of the current generation of PROs. Because the NDI was originally developed as an instrument for patients with neck pain and whiplash-associated disorders, the majority of validation of the NDI has focused on this patient population. 13 The authors could not identify any study that has validated the psychometric properties of NDI in patients with myelopathy, although it has been used to describe outcomes in this patient population. 14,15,16 Furthermore, the literature on NDI in patients with cervical radiculopathy has yielded mixed results. 9,17,18,19,20 For instance Young et al have shown that the NDI has poor construct validity in patients with cervical radiculopathy and may suffer from limited test–retest reliability (intraclass correlation coefficient [ICC] = 0.55). 21 Other authors have suggested that the NDI may be multidimensional instrument making it difficult to calculate change of score and other parametric statistics with NDI data. 20 And although NDI has been shown to be responsive to cervical surgery, there is no clear definition of a minimum clinically important difference (MCID); values calculated in the literature vary from 3.5 to 9.5 on a 0 to 50 scale. 9,11,17,22 The difficulty in calculating MCID and larger MCIDs has been attributed to the fact that the NDI was not designed with cervical radiculopathy patients in mind. 13 The NDI may also have significant floor effects in patients with cervical spine disorders. 23
To our knowledge, there are no studies that have validated or attempted to validate the utility of NDI in patients with cervical deformity. The NDI functions as a pain-interference (PI) scale; that is, it seeks to examine how neck pain affects functions of daily living. Therefore, it may serve to measure how cervical deformity affects neck pain. It does not, however, directly address important consequences of cervical deformity (e.g., in ability to maintain horizontal gaze and difficulty swallowing).
General outcome measures such as the SF-36 also have substantial shortcomings. In a survey of 147 patients, Baron and colleagues found that SF-36 scores in cervical patients did not follow the same patterns as the general population. 24 Additionally, in the cervical population, the two components of the SF-36 score (mental and physical) did not explain as much of the variance in SF-36 scales as required. These authors also found significant floor/ceiling effect in multiple scales of the SF-36 and concluded that reporting SF-36 summary scores in patients with neck disease was inappropriate and misleading. 24
The alternatives to PROs have been surgeon-administered instruments such as the mJOA and Nurick scales. These have undergone only limited psychometric evaluation; validation of the mJOA is limited to a single, recently published study. 25 Although this study described the mJOA as a useful tool in assessing functional status in patients with myelopathy, the outcome measure had only moderate internal consistency (Cronbach α = 0.63) and was multidimensional. The authors also found that this functional measure of disability was poorly correlated with patient-related outcome measures such as the NDI and SF-36. In addition, the authors were not able to evaluate inter- and intra-rater reliability of this surgeon-administered tool. 25 The JOA and mJOA, however, have been used to measure disability in the cervical spine.
Newer PROs such as the National Institutes of Health (NIH) Patient Reported Outcomes Measurement Information System (PROMIS) may have the potential to overcome the aforementioned shortcomings. The PROMIS was developed to allow practitioners to determine global and domain-specific outcome measures; additionally, it uses tools such as computer adaptive testing (CAT) to minimize floor and ceiling effects and reduce questionnaire burden. While there has been some preliminary work performed on the NIH PROMIS in cervical spine patients, there are currently no studies on cervical deformity patients.
In addition to the NIH PROMIS, it might also be useful to devise a disease-specific measure to utilize in patients with cervical deformity. An ideal measure would measure impairment due to limitations in achieving horizontal gaze, range of motion, and dysphagia. There is currently no single outcome measure to address each of these domains; all of which have special relevance to cervical deformity patients.
While the development of such disability measures might represent an interesting area of future research, there have been several recent publications that have examined the impact of cervical malalignment on disability. Cervical malalignment can be considered in one of three different regions: occipitocervical and upper cervical alignment, subaxial cervical alignment, and cervicothoracic alignment.
6.3 Occipitocervical and Upper Cervical Alignment and Measures of Horizontal Gaze
As noted earlier, horizontal gaze is an important requirement for most activities of daily living. The alignment between the occiput and the cervical spine serves as an important regulator of this function. The first occipitocervical parameter to be correlated to disability was the chin–brow vertical angle (CBVA). Suk et al described the significance of the CBVA in a series of ankylosing spondylitis patients. They showed that patients with elevated CBVA had significantly lower scores for horizontal gaze compared to patients with corrected CBVA. 26 Several authors have since commented on the importance of horizontal gaze and other similar parameters have been described that are closely correlated to the CBVA. These include the slope line of sight (SLS) and McGregor slope (McGS). 27,28,29 Although Suk et al found no correlation between CBVA and HRQOL in a modified arthritis scale, 26 other authors have used a regression analysis to suggest optimal targets for CBVA, SLS, and McGS in patients with cervical deformity. Lafage et al used regression analysis in 303 patients presenting to a deformity clinic and determined that CBVA ranging from -4.7 to 17.7 degrees, SLS ranging from -5.1 to 18.5 degrees, and McGS ranging from -5.7 to 14.3 degrees correlated to lower patient disability scores on the Oswestry Disability Index (ODI). 27 It should be noted, however, that the ODI measures lower back dysfunction and may not be broadly applicable to patients with cervical deformity.
Bao et al recently published a principal components analysis of all patients with cervical deformity. 30 Their goal was to determine which components of cervical alignment were mostly correlated to disability. They compared 171 asymptomatic and 107 symptomatic patients and attempted to determine which cervical alignment parameters were best able to discriminate between the two groups. In their analysis, SLS and McGS were different between the symptomatic and asymptomatic cohorts and SLS was an independent predictor of cervical disability. The principal component revealed three principal components: cranial orientation (SLS, McGS, CBVA), occipitocervical orientation (C0–C2), and lower cervical alignment (C2–C7 Cobb, C2–C7 SVA).
The C0–C2 angle is an additional occipitocervical parameter that has been correlated to disability in patients with cervical malalignment. As noted earlier, C0–C2 is an important component of cervical alignment in patients with neck complaints. Izeki et al showed the importance of the C0–C2 angle in setting the oropharyngeal space. In their experience, patients who are fused with a more kyphotic C0–C2 angle were more likely to have a narrowed oropharyngeal space and more likely to experience dysphagia and disability. 31
6.4 Subaxial Cervical Alignment
Historically, “malalignment” in the subaxial cervical spine was considered to mean cervical kyphosis. This focus on cervical kyphosis stems partially from the disability experienced by patients suffering from post-laminectomy kyphosis. 32,33,34,35 More recent data, however, have shown that cervical kyphosis may not necessarily represent “abnormal” alignment. Diebo and colleagues showed that cervical kyphosis might represent a normal alignment profile in a significant number of patients, particularly those with low thoracic kyphosis (TK). 8 Studies of asymptomatic patients have also shown that a significant number (up to 35%) can have cervical kyphosis 36 and additional studies have since shown that cervical curvature has a poor correlation with measures of disability in postoperative patients. 37,38,39
That said, it is important to consider C2–C7 curvature in patients with cervical disability. In the principal components analysis by Bao et al, the C2–C7 angle was an important component of disability when considered in conjunction with upper cervical alignment and the C2–C7 sagittal vertical axis (SVA). 30 Similarly, Iyer et al reported on a series of 90 preoperative patients who were presented to a cervical spine clinic for surgery. In this series, an increased C2–C7 lordosis was correlated to increased disability as measured by the NDI. 40 This finding has since been corroborated in a series of postoperative patients as well. 37