The influence of conventional T2 MRI indices in predicting who will walk outside one year after spinal cord injury

Predictive modeling of ambulatory outcomes after spinal cord injury using machine learning

STUDY DESIGN

Retrospective multi-site cohort study.

OBJECTIVES

To develop an accurate machine learning predictive model using predictor variables from the acute rehabilitation period to determine ambulatory status in spinal cord injury (SCI) one year post injury.

SETTING

Model SCI System (SCIMS) database between January 2000 and May 2019.

METHODS

Retrospective cohort study using data that were previously collected as part of the SCI Model System (SCIMS) database. A total of 4523 patients were analyzed comparing traditional models (van Middendorp and Hicks) compared to machine learning algorithms including Elastic Net Penalized Logistic Regression (ENPLR), Gradient Boosted Machine (GBM), and Artificial Neural Networks (ANN).

RESULTS

Compared with GBM and ANN, ENPLR was determined to be the preferred model based on predictive accuracy metrics, calibration, and variable selection. The primary metric to judge discrimination was the area under the receiver operating characteristic curve (AUC). When compared to the van Middendorp all patients (0.916), ASIA A and D (0.951) and ASIA B and C (0.775) and Hicks all patients (0.89), ASIA A and D (0.934) and ASIA B and C (0.775), ENPLR demonstrated improved AUC for all patients (0.931), ASIA A and D (0.965) ASIA B and C (0.803).

CONCLUSIONS

Utilizing artificial intelligence and machine learning methods are feasible for accurately classifying outcomes in SCI and may provide improved sensitivity in identifying which individuals are less likely to ambulate and may benefit from augmentative strategies, such as neuromodulation. Future directions should include the use of additional variables to further refine these models.

Longitudinal motor system changes from acute to chronic spinal cord injury

BACKGROUND AND PURPOSE

In acute spinal cord injury (SCI), magnetic resonance imaging (MRI) reveals tissue bridges and neurodegeneration for 2 years. This 5-year study aims to track initial lesion changes, subsequent neurodegeneration, and their impact on recovery.

METHODS

This prospective longitudinal study enrolled acute SCI patients and healthy controls who were assessed clinically-and by MRI-regularly from 3 days postinjury up to 60 months. We employed histologically cross-validated quantitative MRI sequences sensitive to volume, myelin, and iron changes, thereby reflecting indirectly processes of neurodegeneration and neuroinflammation. General linear models tracked lesion and remote changes in volume, myelin- and iron-sensitive magnetic resonance indices over 5 years. Associations between lesion, degeneration, and recovery (using the Spinal Cord Independence Measure [SCIM] questionnaire and the International Standards for Neurological Classification of Spinal Cord Injury total motor score) were assessed.

RESULTS

Patients' motor scores improved by an average of 12.86 (95% confidence interval [CI] = 6.70-19.00) points, and SCIM by 26.08 (95% CI = 17.00-35.20) points. Within 3-28 days post-SCI, lesion size decreased by more than two-thirds (3 days: 302.52 ± 185.80 mm2 , 28 days: 76.77 ± 88.62 mm2 ), revealing tissue bridges. Cervical cord and corticospinal tract volumes transiently increased in SCI patients by 5% and 3%, respectively, accompanied by cervical myelin decreases and iron increases. Over time, progressive atrophy was observed in both regions, which was linked to early lesion dynamics. Tissue bridges, reduced swelling, and myelin content decreases were predictive of long-term motor score recovery and improved SCIM score.

CONCLUSIONS

Studying acute changes and their impact on longer follow-up provides insights into SCI trajectory, highlighting the importance of acute intervention while indicating the potential to influence outcomes in the later stages.

Influence of Blood Pressure on Acute Cervical Spinal Cord Injury Without Fracture and Dislocation: Results From a Retrospective Analysis

OBJECTIVE

The objective of this study was to investigate the influence of blood pressure on the severity and functional recovery of patients with acute cervical spinal cord injury (SCI) without fracture and dislocation.

METHODS

A retrospective case control study analyzed the data of 40 patients admitted to our orthopedics department (Beijing Tiantan Hospital, Capital Medical University) from January 2013 to February 2021. They were diagnosed as acute cervical SCI without fracture and dislocation. Gender, age, height, weight, history of hypertension, postinjury American Spinal Injury Association grade, postinjury modified Japanese Orthopaedic Association (mJOA) score, postoperative mJOA score, 1-year follow-up mJOA score, preoperative mean arterial pressure (MAP), intramedullary T2 hyperintensity, and hyponatremia were collected. The patients were divided into groups and subgroups based on their history of hypertension and preoperative MAP. The effects of history of hypertension and preoperative MAP on the incidence of T2 hyperintensity, hyponatremia, the improvement rate of the postoperative mJOA and 1-year follow-up mJOA scores were analyzed.

RESULTS

Patients with history of hypertension had a lower incidence of intramedullary T2 hyperintensity than patients without history of hypertension (P < 0.05). Patients with history of hypertension and patients with a higher preoperative MAP had better neurological recovery at 1 year of follow-up (P < 0.05).

CONCLUSIONS

Blood pressure has great influence on acute cervical SCI without fracture and dislocation. Maintaining a higher preoperative MAP is advantageous for better recovery after SCI. Attention should be paid to the dynamic management of blood pressure to avoid the adverse effects of hypotension after SCI.

Peak Slope Ratio of the Recruitment Curves Compared to Muscle Evoked Potentials to Optimize Standing Configurations with Percutaneous Epidural Stimulation after Spinal Cord Injury

Background: Percutaneous spinal cord epidural stimulation (pSCES) has effectively restored varying levels of motor control in persons with motor complete spinal cord injury (SCI). Studying and standardizing the pSCES configurations may yield specific motor improvements. Previously, reliance on the amplitude of the SCES-evoked potentials (EPs) was used to determine the correct stimulation configurations. Methods: We, hereby, retrospectively examined the effects of wide and narrow-field configurations on establishing the motor recruitment curves of motor units of three different agonist-antagonist muscle groups. Magnetic resonance imaging was also used to individualize SCI participants (n = 4) according to their lesion characteristics. The slope of the recruitment curves using a six-degree polynomial function was calculated to derive the slope ratio for the agonist-antagonist muscle groups responsible for standing. Results: Axial damage ratios of the spinal cord ranged from 0.80 to 0.92, indicating at least some level of supraspinal connectivity for all participants. Despite the close range of these ratios, standing motor performance was enhanced using different stimulation configurations in the four persons with SCI. A slope ratio of ≥1 was considered for the recommended configurations necessary to achieve standing. The retrospectively identified configurations using the supine slope ratio of the recruitment curves of the motor units agreed with that visually inspected muscle EPs amplitude of the extensor relative to the flexor muscles in two of the four participants. Two participants managed to advance the selected configurations into independent standing performance after using tonic stimulation. The other two participants required different levels of assistance to attain standing performance. Conclusions: The findings suggest that the peak slope ratio of the muscle agonists-antagonists recruitment curves may potentially identify the pSCES configurations necessary to achieve standing in persons with SCI.

A Single Dermatome Clinical Prediction Rule for Independent Walking 1 Year After Spinal Cord Injury

OBJECTIVE

To derive and validate a simple, accurate CPR to predict future independent walking ability after SCI at the bedside that does not rely on motor scores and is predictive for those initially classified in the middle of the SCI severity spectrum.

DESIGN

Retrospective cohort study. Binary variables were derived, indicating degrees of sensation to evaluate predictive value of pinprick and light touch variables across dermatomes. The optimal single sensory modality and dermatome was used to derive our CPR, which was validated on an independent dataset.

SETTING

Analysis of SCI Model Systems dataset.

PARTICIPANTS

Individuals with traumatic SCI. The data of 3679 participants (N=3679) were included with 623 participants comprising the derivation dataset and 3056 comprising the validation dataset.

INTERVENTIONS

Not applicable.

MAIN OUTCOME MEASURES

Self-reported ability to walk both indoors and outdoors.

RESULTS

Pinprick testing at S1 over lateral heels, within 31 days of SCI, accurately identified future independent walkers 1 year after SCI. Normal pinprick in both lateral heels provided good prognosis, any pinprick sensation in either lateral heel provided fair prognosis, and no sensation provided poor prognosis. This CPR performed satisfactorily in the middle SCI severity subgroup.

CONCLUSIONS

In this large multi-site study, we derived and validated a simple, accurate CPR using only pinprick sensory testing at lateral heels that predicts future independent walking after SCI.

The presence or absence of midsagittal tissue bridges and walking: a retrospective cohort study in spinal cord injury

STUDY DESIGN

Cohort study. Retrospective analysis of T2-weighted magnetic resonance images (MRIs) and clinical documentation.

OBJECTIVES

To evaluate the relationship between the presence/absence and widths of midsagittal tissue bridges and walking ability among veterans with cervical, predominantly chronic SCI.

SETTING

University research and hospital setting.

METHODS

T2-weighted midsagittal MRIs of 22 United States veterans with cervical spinal cord injuries were examined. The presence/absence of midsagittal tissue bridges were determined, and the widths of present ventral and dorsal tissue bridges were measured. Midsagittal tissue bridge characteristics were related to each participant's ability to walk based off examination of clinical documentation.

RESULTS

Fourteen of the analyzed participant images revealed the presence of midsagittal tissue bridges. Ten of those individuals (71%) possessed overground walking ability. The 8 individuals with no apparent tissue bridges were all unable to walk. There was a significant correlation between walking and widths of ventral midsagittal tissue bridges (r = 0.69, 95%CI: 0.52, 0.92, p < 0.001), as well as dorsal midsagittal tissue bridges (r = 0.44, 95%CI: 0.15, 0.73, p = 0.039).

CONCLUSION

The evaluation of midsagittal tissue bridges may be useful in various rehabilitation settings to help inform patients' plan of care, allocation of neuromodulatory resources, and appropriate stratification into research cohorts.

Transcutaneous Electrical Spinal Cord Stimulation to Promote Recovery in Chronic Spinal Cord Injury

Objective: To evaluate the impact of using transcutaneous electrical spinal cord stimulation (TSCSTSCS) on upper and lower extremity function in individuals with chronic spinal cord injury (SCI).

Design: Prospective case series.

Setting: SCI specific rehabilitation hospital.

Participants: A convenience sample (N = 7) of individuals with tetraplegia who had previously been discharged from outpatient therapy due to a plateau in progress.

Interventions: Individuals participated in 60 min of upper extremity (UE) functional task-specific practice (FTP) in combination with TSCS and 60 min of locomotor training in combination with TSCS 5x/week.

Main Outcome Measures: The primary outcome for this analysis was the Capabilities of Upper Extremity Test (CUE-T). Secondary outcomes include UE motor score (UEMS), LE motor score (LEMS), sensation (light touch and pin prick), Nine-Hole Peg Test, 10 meter walk test, 6 min walk test, and 5 min stand test.

Results: Seven individuals (four motor complete; three motor incomplete) completed 20–80 sessions UE and LE training augmented with TSCS and without any serious adverse events. Improvements were reported on the CUE-T in all seven individuals. Two individuals improved their ASIA impairment scale (AIS) classification (B to C; C to D) and two individuals improved their neurologic level of injury by one level (C4–C5; C5–C6). Sensation improved in five individuals and all four who started out with motor complete SCIs were able to voluntarily activate their LEs on command in the presence of stimulation.

Conclusion: Individuals with chronic SCI who had previously demonstrated a plateau in function after an intensive outpatient therapy program were able to improve in a variety of UE and LE outcomes in response to TSCS without any adverse events. This was a small pilot study and future fully powered studies with comparative interventions need to be completed to assess efficacy.

Lateral Corticospinal Tract and Dorsal Column Damage: Predictive Relationships With Motor and Sensory Scores at Discharge From Acute Rehabilitation After Spinal Cord Injury

OBJECTIVE

To determine if lateral corticospinal tract (LCST) integrity demonstrates a significant predictive relationship with future ipsilateral lower extremity motor function (LEMS) and if dorsal column (DC) integrity demonstrates a significant predictive relationship with future light touch (LT) sensory function post spinal cord injury (SCI) at time of discharge from inpatient rehabilitation.

DESIGN

Retrospective analyses of imaging and clinical outcomes.

SETTING

University and academic hospital.

PARTICIPANTS

A total of 151 participants (N=151) with SCI.

INTERVENTIONS

Inpatient rehabilitation.

MAIN OUTCOME MEASURES

LEMS and LT scores at discharge from inpatient rehabilitation.

RESULTS

In 151 participants, right LCST spared tissue demonstrated a significant predictive relationship with right LEMS percentage recovered (β=0.56; 95% confidence interval [CI], 0.37-0.73; R=0.43; P<.001). Left LCST spared tissue demonstrated a significant predictive relationship with left LEMS percentage recovered (β=0.66; 95% CI, 0.50-0.82; R=0.51; P<.001). DC spared tissue demonstrated a significant predictive relationship with LT percentage recovered (β=0.69; 95% CI, 0.52-0.87; R=0.55; P<.001). When subgrouping the participants into motor complete vs incomplete SCI, motor relationships were no longer significant, but the sensory relationship remained significant. Those who had no voluntary motor function but recovered some also had significantly greater LCST spared tissue than those who did not recover motor function.

CONCLUSIONS

LCST demonstrated significant moderate predictive relationships with lower extremity motor function at the time of discharge from inpatient rehabilitation, in an ipsilesional manner. DC integrity demonstrated a significant moderate predictive relationship with recovered function of LT. With further development, these neuroimaging methods might be used to predict potential deficits after SCI and to provide corresponding targeted interventions.

Spinal cord pathology revealed by MRI in traumatic spinal cord injury

PURPOSE OF REVIEW

This review covers recent advances in identifying conventional and quantitative neuroimaging spinal cord biomarkers of lesion severity and remote spinal cord pathology following traumatic spinal cord injury (SCI). It discusses the potential of the most sensitive neuroimaging spinal cord biomarkers to complement clinical workup and improve prediction of recovery.

RECENT FINDINGS

At the injury site, preserved midsagittal tissue bridges - based on conventional sagittal T2-weighted scans - can be identified in the majority of SCI patients; its width being predictive of recovery. Remote from the injury, diffusion indices, and myelin/iron-sensitive neuroimaging-based changes are sensitive to secondary disease processes; its magnitude of change being associated with neurological outcome.

SUMMARY

Neuroimaging biomarkers reveal focal and remote cord pathology. These biomarkers show sensitivity to the underlying disease processes and are clinically eloquent. Thus, they improve injury characterization, enable spatiotemporal tracking of cord pathology, and predict recovery of function following traumatic SCI. Neuroimaging biomarkers, therefore, hold potential to complement the clinical diagnostic workup, improve patient stratification, and can serve as potential endpoints in clinical trials.