Diffusion tensor imaging as an additional postoperative prognostic predictor factor in cervical myelopathy patients: An observational study.

INTRODUCTION: Multiple investigation modalities have been invented for diagnosis and for planning management of degenerative cervical myelopathy, which include magnetic resonance imaging (MRI), computed tomography scan, and plain X-rays. Diffusion tensor imaging (DTI) of the spinal cord is a special variety of MRI where diffusion of water molecules across and along the tracts is mapped. The changes in anisotropy at the stenotic level can be a postoperative prognostic factor. The aim of this study was to establish postoperative prognostic predictive value of DTI in cases of degenerative cervical myelopathy.
MATERIALS AND METHODS: The study included 30 indoor patients in a tertiary care hospital diagnosed with degenerative compressive cervical myelopathy based on both clinical and radiological parameters with complete clinical data including follow-up. All patients with medical neurological diseases, cases who underwent repeat surgery, cases who developed surgical site infection, and those patients who were lost to follow-up were excluded from the study. The patients underwent operative decompression through either anterior or posterior approach with or without fixation with titanium implants as per indication. All patients underwent pre- and postoperative DTI. The fractional anisotropy (FA) and apparent diffusion coefficient (ADC) were noted in both pre- and postoperative imaging. Epidemiological data such as age and sex were noted. Pre- and postoperative modified Japanese Orthopedic Association (mJOA) scores were calculated.
RESULTS: There was a significant improvement in FA values postoperatively. Preoperatively, both FA and ADC values showed a significant correlation with preoperative Neurological status of the patient while postoperatively only FA values were found to be significantly correlated. The regression equations for determining postoperative mJOA score based on preoperative FA and ADC values revealed mJOA = 9.77 + 12.1 (FA), mJOA = 14.2 + 2408.4 (ADC), and mJOA = 9.54 + 11.2 (FA) +1575.5 (ADC). This means that postoperative mJOA score, i.e., postoperative clinical status improvement can be determined using DTI variables which are an objective preoperative data. However, relative strength of prediction for FA value is 66.7% and for ADC value is 28.7%.
CONCLUSION: DTI tractography of the spinal cord will be a helpful objective prognostic factor for patients in whom surgery is planned. However, a study with larger subject size is required to increase the accuracy of determination of regression coefficient.

INTRODUCTION

Multiple investigation modalities have been invented for diagnosis and for planning management of degenerative cervical myelopathy, which include magnetic resonance imaging (MRI), computed tomography scan, and plain X-rays. Diffusion tensor imaging (DTI) of the spinal cord is a special variety of MRI where diffusion of water molecules across and along the tracts is mapped. The changes in anisotropy at the stenotic level can be a postoperative prognostic factor.

The aim of this study was to establish postoperative prognostic predictive value of DTI in cases of degenerative cervical myelopathy.

MATERIALS AND METHODS

The study included 30 indoor patients (29 males and 1 female) who visited a tertiary care hospital during a period from September 2013 to September 2015 diagnosed with degenerative compressive cervical myelopathy based on both clinical and radiological parameters with complete clinical data including follow-up. All patients with medical neurological diseases, cases who underwent repeat surgery, cases who developed surgical site infection, and those patients who were lost to follow-up were excluded from the study. The patients underwent operative decompression through either anterior or posterior approach with or without fixation with titanium implants as per indication.

After approval from the Institutional Ethics Committee, all patients underwent pre- and postoperative DTI of the cervical spine with 1.5-Tesla magnetic power imaging machine [Figures 1 and 2]. Postoperative imaging was done 6 weeks postoperatively after complete wound healing and subsidence of soft-tissue edema.

Figure 1

Preoperative diffusion tensor imaging–magnetic resonance imaging

Figure 2

Preoperative diffusion tensor–imaging magnetic resonance imaging (magnified view)

The fractional anisotropy (FA) and apparent diffusion coefficient (ADC) were noted in both pre- and postoperative imaging. Epidemiological data such as age and sex were noted. Pre- and postoperative modified Japanese Orthopedic Association (mJOA) scores were calculated [Figure 3].

Figure 3

Modified Japanese Orthopedic Association score

RESULTS

There was a significant improvement in FA values postoperatively. Preoperatively, both FA and ADC values showed a significant correlation with preoperative neurostatus of the patient. This means that, on preoperative imaging, the reduction in FA and increase in transverse ADC at the stenotic level could be positively correlated with severity judged by mJOA score [Tables 1 and 2], while postoperatively only improvement in FA values could be correlated significantly with improved mJOA score. The regression equations for determining postoperative mJOA score based on preoperative FA and ADC values revealed:

Table 1

Regression Model (mJOA post op as Dependent Variable)

Variables	Unstandardized coefficient		Standardized co efficient	t	P	95% co efficient interval for B
	B	Std. error	Beta			Lower bound	Upper bound
(Constant)	9.77	1.00		9.72	<0.01	7.71	11.83
FA Pre-op	12.10	2.13	0.73	5.69	<0.01	7.74	16.46

Table 2

Regression Model (mJOA post op as Dependent Variable)

Variables	Unstandardized coefficient		Standardised co efficient	t	P	95% co efficient interval for B
	B	Std. Error	Beta			Lower bound	Upper bound
(Constant)	9.547	0.938		10.174	<0.01	7.621	11.472
FA Pre-op	11.028	2.03	0.667	5.432	<0.01	6.862	15.193
ADC pre op	1575.518	675.177	0.287	2.333	0.027	190.169	2960.867

mJOA = 9.77 + 12.1 (FA)

mJOA = 14.2 + 2408.4 (ADC)

mJOA = 9.54 + 11.2 (FA) +1575.5 (ADC).

This means that postoperative mJOA score, i.e., postoperative clinical status improvement can be determined using DTI variables which are an objective preoperative data. However, relative strength of prediction for FA value is 66.7% and for ADC value is 28.7%. This means that, in only 66.7% of cases, postoperative mJOA score could be accurately predicted based on preoperative FA values. It was also found that in severe cases (mJOA <9), the predicted postoperative improved score exceeds actual score, but statistical significance of this finding could not be established because of smaller subject size.

DISCUSSION

The degenerative compressive cervical myelopathy is one of the common disabling conditions affecting the cervical spine in adults. The pathologies include degenerative disc disease, ossified posterior longitudinal ligament, and ligamentum flavum hypertrophy with or without associated instability.[1234]

The basic motivation for this study was a common clinical to imaging mismatch arising out of variable individual tolerance of the spinal cord to the compression caused by degenerative spinal canal narrowing. It has been demonstrated by animal model experiments that spinal cord function may be surprisingly resistant to compression. This also manifests in the relatively common accidental findings of asymptomatic spondylotic cervical spinal cord compression. Thorpe et al. described degenerative changes of the cervical spine in 64% of a target group of asymptomatic individuals at ages between 18 and 72 years, and spinal cord compression was noted in 11% of the group.

Conventional MRI[567] is routinely used to image the spinal cord to demonstrate the location and amount of cord compression from extramedullary factors such as prolapsed disc, ossification of ligamentum flavum, or posterior longitudinal ligament. However, apart from the instability factor demonstrated by dynamic X-ray imaging,[8] signal intensity changes seen in conventional MRI[9] failed to demonstrate any prognostic value.

The DTI of spinal cord studies diffusion of water molecules across and along the tracts. Displaced water molecules produce an attenuated signal during diffusion MR scanning. By its nature, the axonal architecture in CNS white matter promotes diffusion of water molecules in a direction predominantly parallel, rather than perpendicular, to axon fibers. Diffusion perpendicular to the fibers seems to be limited by cell membranes more than myelin sheaths. This direction-dependent diffusion is described as “anisotropy” and is used by DTI to infer the orientation of surrounding axonal fibers and to delineate anatomical boundaries. DTI uses a tensor framework to characterize molecular motion in multiple directions in a three-dimensional space. The diffusivities along the three principal axes are defined by the eigenvectors where λ1 (primary eigenvector) represents the direction and magnitude of the longitudinal diffusion vector, while λ2 and λ3 represent vectors along the minor axes. The magnitudes of these vectors are used to calculate a number of indices, of which the commonly used parameters are FA, ADC transverse and longitudinal, and mean diffusivity. FA ranges from 0 to 1, which defines the degree of anisotropy, and tissues with high anisotropy have a value closer to 1. The eigenvalues are affected by microstructural alterations that affect the diffusion of water molecules, and this forms the basis for using DTI indices to identify spinal cord pathology.

Although it was initially done for spinal cord injuries, tumors, and neurological diseases such as multiple sclerosis,[91011] degenerative compressive myelopathy also demonstrated significant changes in FA and ADC values of the affected stenotic segment in both animal models and human beings.[12]

The advantage of this imaging is that it gives objective data, which removes interobserver variability in interpretation of the conventional MRI. In various studies, it was established that there is a significant decrease in FA and ADC values at the stenotic level of the cervical spine. In our study, we found that there was an improvement in these values after operative decompression. The pre- and postoperative clinical status of these patients was assessed using mJOA. The correlation between preoperative values of FA and ADC with postoperative mJOA score was established using statistical analysis.

This study revealed that there was a significant correlation between preoperative FA and ADC values measured at the level of maximum compression (as seen on conventional imaging) with preoperative mJOA score, that is, preoperative clinical status of the patient can be predicted using these values. The regression equations for preoperative FA and ADC values and postoperative mJOA score were established during statistical analysis of this study. These equations can be used to predict postoperative mJOA score for a particular patient using preoperative FA values. However, as stated earlier, the relative strength of prediction of accurate value using FA was only 66.7% and that for ADC was 28.7%, which means that a study with larger subject size is required to increase accuracy of prediction and to establish more accurate equation. It should also be noted that the relative reduction in the anisotropy would be more relevant than the absolute values of FA and ADC in predicting the outcome. The accuracy can also be improved using 3-Tesla MRI technique.

CONCLUSION

DTI of the spinal cord will be a helpful objective prognostic factor for patients in whom surgery is planned. However, a study with larger subject size is required to increase the accuracy of determination of regression coefficient. The cost–benefit ratio still remains a question in Indian scenario.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.