Implementing a rapid cord compression Magnetic Resonance Imaging protocol in the emergency department: Lessons learned

Abstract

Background & Purpose

(1) Evaluate efficacy of an abbreviated total spine protocol in triaging emergency department (ED) patients through retrospective evaluation. (2) Describe patient outcomes following implementation of a rapid cord compression protocol.

Methods

(1) All contrast-enhanced total spine magnetic resonance imaging studies (MRIs) performed on ED patients (n = 75) between 10/1–12/31/2022 for evaluation of cord compression were included. Two readers with 6 and 5 years of experience blindly reviewed the abbreviated protocol (comprised of sagittal T2w and axial T2w sequences) assessing presence of cord compression or severe spinal canal stenosis. Ground truth was consensus by a neuroradiology fellow and 2 attendings. (2) The implemented rapid protocol included sagittal T1w, sagittal T2w Dixon and axial T2w images. All ED patients (n = 85) who were imaged using the rapid protocol from 5/1–8/31/2023 were included. Patient outcomes and call-back rates were determined through chart review.

Results

(1) Sensitivity and specificity for severe spinal canal stenosis and/or cord compression was 1.0 and 0.92, respectively, for reader 1 and 0.78 and 0.85, respectively, for reader 2. Negative predictive value was 1.0 and 0.97 for readers 1 and 2, respectively. (2) The implemented rapid cord compression protocol resulted in 60% reduction in imaging time at 1.5T. The call-back rate for additional sequences was 7%. In patients who underwent surgery, no additional MRI images were acquired in 82% of cases (9/11).

Conclusions

Implementing an abbreviated non-contrast total spine protocol in the ED results in a low call-back rate with acquired MRI images proving sufficient for both triage and treatment planning in most patients.

Introduction

Utilization of advanced imaging in the emergency department (ED), particularly magnetic resonance imaging (MRI) has been substantially rising over the last two decades. ¹ Moreover, increasing utilization of MRI and other advanced imaging in the ED has been found to significantly contribute to length of stay and time to discharge of ED admissions. ² While there has been increasing scrutiny by governmental and insurance agencies on the utilization of resource-intensive imaging studies, MRI of the spine is the sole modality which allows diagnostic evaluation of the spinal cord and cauda equina nerve roots. Patient history and physical exam is often inadequate in the diagnosis and localization of acute spinal pathology, particularly in the setting of spinal cord compression ³ or cauda equina syndrome, ⁴ where delayed diagnosis and treatment may result in significant morbidity. ³ Therefore, in patients presenting with acute back pain and acute neurological deficits, total spine MRI evaluation is often indicated. ⁵

Previous studies evaluating the yield of total MRI spine evaluations in the ED have found major radiologic findings in up to 36.8% of patients, of whom nearly half underwent some treatment within 24 hours. ⁶ Prior studies focused on imaging of single spinal levels rather than total spine MRI also demonstrated similar findings with major radiologic findings found in 33%–58% of patients with up to 36.8% undergoing treatment within 48 hours. ⁷ Despite the demonstrated clinical utility of emergent spinal imaging evaluation with MRI, long imaging times contribute to increased wait-times and length-of-stay and diminished patient throughput. ²

To combat long imaging times, rapid and abbreviated MRI protocols have increasingly been developed as alternatives to standard MRI protocols with reduced numbers of sequences and imaging time to answer targeted clinical questions, particularly in the emergency department setting. ⁸ These have most commonly been employed for imaging of acute ischemic stroke in adult patients ⁹ and evaluation of shunted hydrocephalus in pediatric patients. ¹⁰ For spinal imaging in the ED, use of an abbreviated lumbar MRI protocol has been reported using a rapid 3D-T2w Sampling Perfection with Application optimized Contrasts using different flip angle Evolution (3D T2 SPACE) fat-saturated (FS) sequence specifically for screening evaluation in patients presenting with acute low back pain. The authors found that the 3D T2w SPACE FS sequence was an effective screen for fracture, cord signal abnormality, and severe spinal canal stenosis with both sensitivities and specificities between 96%–100%. ¹¹

Evaluation of an abbreviated total spine protocol for assessment of cord compression has been previously reported in the literature by one group who retrospectively reviewed abbreviated protocol images inclusive of Short Tau Inversion Recovery (STIR), and axial T2w images of the total spine. In this non-inferiority, retrospective study, the authors reported 100% sensitivity, 98.6% specificity for detection of spinal cord compression or cauda equina nerve compression, and 100% sensitivity, 99.3% specificity for detection of other clinically relevant findings, inclusive of lesions in the osseous structures, spinal cord, or epidural space. ¹²

In this study, we aim to extend the existing literature through a two-part study. In the first phase of this study, we performed a retrospective study evaluating the sensitivity, specificity, positive predictive value, and negative predictive value of a proposed abbreviated protocol consisting of sagittal and axial T2w images. Following results of this retrospective evaluation, consensus between the radiology, emergency medicine, and neurosurgery departments were made to begin implementation of an abbreviated non-contrast total spine protocol for triage of emergency department patients. Here, we report initial results following implementation of such a protocol for triage of emergency department patients suspected of spinal cord compression in the first 3 months following implementation, through review of the electronic medical record.

Methods

Retrospective phase

Imaging protocol

Following institutional review board approval for this retrospective study with waiver of consent, all contrast-enhanced total spine MRIs performed on ED patients (n = 75) between October 1^st and December 31^st, 2022 for evaluation of cord compression were identified, reviewed, and included if the protocol was completed in entirety. All were imaged on 1.5T magnets, either Siemens Sola 1.5T (Siemens Healthineers, Erlangen, Germany) or Phillips 1.5T (Achieva or Ingenia; Phillips Medical Systems, Best, The Netherlands), with approximate protocol lengths 45:56 and 67:49 (in minutes:seconds) on Siemens and Phillips 1.5T, respectively. Exact protocol lengths were patient-dependent based on the height of the patient. Sequence parameters are included for Siemens Sola 1.5T in Table 1 and Phillips 1.5T in Supplemental Table 1. Proposed sequences for the abbreviated protocol included just the sagittal and axial T2w image stacks.

Table 1.

Imaging protocol and parameters for the baseline total spine protocol without and with intravenous contrast on Siemens 1.5T. All sequences were performed twice with top stack images covering skull base to T6/T7 level and bottom stack images including T6/T7 to mid sacrum. Sag = Sagittal. Ax = Axial. +C = post-contrast. STIR = Short Tau Inversion Recovery.

Sequence type	Sag T2w	Sag STIR	Sag T1w	AxiT2w	Axial T1w	Axial T1 Dixon+C	Sag T1 Dixon+C
Repetition time (ms)	3500	4000	553	5400	650	644	600
Echo time (ms)	99	39	10	96	10	11	13
Flip angle	150	150	150	150	150	148	150
Turbo factor	19	14	3	17	4	3	4
Echo spacing (ms)	11.0	9.82	10.1	12.0	10.5	11.06	12.60
Bandwidth (Hz/Px)	199	189	219	181	206	501	284
Concatenations	1	1	1	2	4	3	2
Inversion time (ms)	NA	160	NA	NA	NA	NA	NA
Averages (nex)	2	2	3	1	1	2	1
GRAPPA acceleration factor	2	2	2	2	2	2	2
Field of view readout (mm)	340	340	340	180	180	180	340
Matrix size	448 × 358	400 × 320	400 × 300	304 × 228	256 × 192	256 × 192	400 × 240
Slice thickness (mm)	4	4	4	5	5	5	4
Acquisition time (min:sec)	2:18	2:54	2:31	3:59	3:19	4:11	3:46

Imaging review

Two readers with 6 and 5 years of professional experience each reviewed the abbreviated protocol images (inclusive of the sagittal and axial T2w image stacks) blinded to additional sequences and patient history. Readers assessed images on Sectra Research Picture Archiving and Communication System (PACS) (Sectra, Inc., Linkoping, Sweden) and provided responses to the following questions:

• (1) Is there severe spinal cord compression with or without cord edema at any level? Respond with Yes, No, Cannot be Determined (i.e., due to factors such as motion and poor image quality)

• (2) Is there severe spinal canal stenosis that may necessitate urgent neurosurgical consult/intervention at any level? Yes, No, Cannot be determined.

• (3) Are there findings that may necessitate call-back for further characterization (specifically T1w or contrast-enhanced imaging)?

Data analysis

For analysis, the answers to question 1 and question 2 were aggregated such that a response of yes to either question 1 or question 2 was considered positive and answer of no to both questions 1 and 2 was considered negative. The reason was because, for the purposes of this analysis, both acute cord compression and cauda equina syndrome were considered positive. The ground truth was consensus evaluation of all sequences by two attending neuroradiologists with over 20 years of professional experience each with a neuroradiology fellow. The neuroradiology fellow performing consensus evaluation was one of the two readers. However, there was approximately a 4-month interval between consensus evaluation and when she served as the reader, and images were randomized and presented on a separate research PACS system when she served as a reviewer. Sensitivity, specificity, positive predictive value, and negative predictive value were calculated for each reader individually. Additionally, inter-reader agreement on responses to the three questions was assessed with Cohen’s kappa coefficients.

Prospective phase

Imaging protocol

Following divisional consensus, the final rapid cord compression protocol imaging protocol is shown in Table 2 for the Siemens 1.5T scanner only. A fixed, larger field of view (FOV) in the phase encoding direction was employed in the sagittal sequences to diminish need for patient-specific FOV adjustment and save positioning time by the performing technologist. This resulted in greater phase encoding steps and slightly longer acquisition times compared to the base protocol. The sagittal T2w sequence was switched to a multi-echo T2w Dixon turbo spin echo to allow for simultaneous acquisition of fat-suppressed images. Similar modifications were made to the imaging protocol for Phillips 1.5T which is shown in Supplemental Table 2. While the initial proposal did not include a sagittal T1w sequence, this was included after divisional consensus because inclusion did not alter the minimum grid timing of the proposed abbreviated MRI.

Table 2.

Imaging protocol and parameters for the implemented rapid cord compression protocol on Siemens 1.5T. All sequences were performed twice with top stack images covering skull base to T6/T7 level and bottom stack images including T6/T7 to mid sacrum. Sag = Sagittal.

Sequence type	Sag T2 Dixon	Sag T1w	Axial T2w
Repetition time (ms)	3030	400	5400
Echo time (ms)	92	10	96
Flip angle	150	150	150
Turbo factor	16	4	17
Echo spacing (ms)	13.1	10.3	12.0
Bandwidth (Hz/Px)	260	219	181
Concatenations	1	2	2
Inversion time (ms)	NA	NA	NA
Averages (nex)	1	2	1
GRAPPA acceleration (acceleration factor, phase encoding)	2	2	2
Field of view readout (mm)	340	340	180
Matrix size	448 × 426	448 × 426	304 × 228
Slice thickness (mm)	4	4	5
Acquisition time (min:sec)	2:52	2:54	3:59

The Siemens 0.55T Free.Max system (Siemens Healthineers, Erlangen, Germany) was acquired in the interval between the two phases of the study. Imaging of emergency department patients on this scanner reflected new institutional work-flow. Patients who were imaged on the Siemens 0.55T system underwent sagittal STIR and T2w turbo spin echo (TSE) rather than sagittal T2w Dixon sequence due to insufficient signal-to-noise ratio on the T2w Dixon images on the 0.55T system (full protocol in Supplemental Table 3). While no included patients in the retrospective phase were imaged with the full total spine protocol on the Siemens 0.55T system, sequence parameters for the total spine protocol without and with intravenous contrast is provided for comparison in Supplemental Table 4.

Imaging review

All ED patients who underwent total spinal imaging from 5/1/2023 to 8/31/2023 were reviewed (n = 113). Patients who were unable to complete the protocol or in cases where additional sequences were specifically requested from the ordering provider (most frequently diffusion weighted imaging) were eliminated. After exclusion, a total of n = 85 patients were included, all imaged on Phillips or Siemens 1.5T (n = 77) or Siemens 0.55 T systems (n = 8). Patients who underwent sagittal STIR and T2w turbo spin echo (TSE) rather than Sagittal T2 Dixon images were included since the STIR images provided similar information to the fat subtracted Dixon images for assessment of soft tissue edema. Call-back rates and patient outcomes of positive cases were determined through review of the finalized report and the electronic medical record.

Results

Retrospective phase

Results are summarized in Table 3. For R1, sensitivity was 1.0 and specificity was 0.92. For R2, sensitivity was 0.78 and specificity was 0.85. For R1, positive predictive value was 0.64 and negative predictive value was 1.0. For R2, positive predictive value was 0.41 and negative predictive value was 0.97.

Table 3.

Sensitivity, specificity, positive predictive value and negative predictive value of each reader (with 5 and 6 years professional experience, respectively) for severe spinal canal stenosis and/or cord compression on retrospective review of axial and sagittal T2w images only. Ground truth was consensus evaluation between a neuroradiology fellow and two neuroradiology attendings.

	Reader 1	Reader 2
Sensitivity	1.0	0.78
Specificity	0.92	0.85
Positive predictive value	0.64	0.41
Negative predictive value	1.0	0.97

Inter-reader agreement for answer to Q1 (presence of cord compression) yielded Cohen’s kappa 0.45 (moderate agreement). Inter-reader agreement for answer to Q2 (presence of severe spinal canal stenosis) yielded Cohen’s kappa 0.68 (substantial agreement). Inter-reader agreement on call-back for additional imaging was 0.64 (substantial agreement).

Prospective phase

Approximate protocol length for the rapid cord compression protocol were 19:30 on the Siemens 1.5T, 24:46 on the Phillips 1.5T, and 41:53 for the Siemens 0.55T (all units are minutes:seconds). These correspond to 58%, 63%, and 52% reduction in imaging time, respectively, when compared to the total cord compression protocol on the same magnets. Of the 85 patients that underwent the rapid cord compression protocol between May 1^st, 2023 and August 31^st, 2023, 55/85 (65%) were rated as negative for cord compression and/or severe spinal canal stenosis (based on final radiology report assessment). Of the remaining, 26/30 had severe spinal canal stenosis, cord compression or concern for nerve root impingement. Four examples of such patients are shown in Figures 1–4. Etiologies of severe spinal canal stenosis or nerve impingement included trauma, metastatic disease, and degenerative disc disease. The remaining 4 studies reported other clinically significant findings inclusive of multifocal areas of T2 signal abnormality in the cord (later found to represent HIV vacuolopathy), concern for arachnoid cyst versus web, nodular intradural lesions (later found to reflect ependymoma recurrence), new osseous metastatic disease without cord compression. Of the 26 studies positive for cord or nerve root compression and/or severe spinal canal stenosis, 11 underwent operative management during the same admission and 15 were either discharged or admitted for other ongoing medical issues. Of the 11 studies that underwent urgent operative management, repeat MRI of the dedicated spinal segment was performed in 2/11 studies for pre-surgical planning purposes with post-contrast images obtained in 1 of the 2 repeated studies. In the remaining 9/11 studies, no subsequent MRI was required prior to operative management.

Figure 2.

32M presenting after 10-foot fall while cutting branches. CT cervical spine was negative for fracture or traumatic malalignment. Rapid cord compression MRI upper stack sagittal T1w (A) and sagittal T2 Dixon (B) images show extensive posterior paraspinal soft tissue and interspinous ligamentous edema. Traumatic disc protrusion with superimposed epidural fluid/hemorrhage contributes to focal severe spinal stenosis at C5-C6 (red arrow). Prevertebral fluid and edema without overt disruption of the anterior longitudinal ligament is also present. Patient underwent urgent anterior cervical disc fusion at C5-C6 without further MR imaging. Axial images not shown.

Figure 3.

84M presenting with known osseous metastatic disease presenting with concern for cord compression. Rapid cord compression MRI lower stack sagittal T1w (A), sagittal T2w (B) images and axial T2w image (C) shows osseous metastasis at L4 with epidural extension resulting in asymmetric narrowing of the right thecal sac with impingement of traversing right-sided nerve roots (red arrow). While imaging was stable from prior, patient’s acute presentation of right lower extremity weakness and radicular pain resulted in urgent L3-L4 decompression, L4 corpectomy and resection of epidural mass with L2-S1 posterior fusion.

Figure 1.

63M presenting after fall from roof with CT showing complete burst fracture of L1 (red arrow) presenting for cord compression MRI. Sagittal T2 Dixon (A), T1w images (B), and axial T2w image (C) show retropulsion of fracture fragments, epidural hemorrhage contributing to cord compression and associated cord edema. Focus of intramedullary low T2 signal (orange arrow) was concerning for intramedullary hemorrhage. The patient underwent subsequent T12-L1 decompression, T11-L3 posterior instrumented fusion without additional MR imaging.

Figure 4.

64M presenting for concern for cord compression. Rapid cord compression MRI cropped lower stack sagittal T1w (A) and sagittal T2w (B) images and axial T2w image (C) show evidence of L2-L4 posterior fusion (partially visualized on these images) with severe spinal canal stenosis with cord compression at T11-T12 (red arrow) contributed by a broad-based disc protrusion, ligamentum flavum thickening and facet hypertrophy. Patient underwent urgent T11-T12 laminectomy without further MRI.

Call-back for additional sequences and/or contrast enhancement was recommended in 6/85 cases (n = 5 on a 1.5T system, n = 1 on a 0.55T system) by the interpreting radiologist. The reason for call-back was for further characterization of a potential marrow-related abnormality in 5/6 cases and for further characterization of an intraspinal mass in 1/6. Additional details and clinical follow-up for call-back cases is discussed below:

• 1. Compression fracture at T7 without severe spinal canal stenosis with neoplastic etiology not excluded. A contrast-enhanced MRI was recommended and completed which did not change management.

• 2. Extensive osseous metastases with soft tissue mass at S3 resulting in obliteration of the inferior thecal sac. Contrast-enhanced total spine MRI was recommended but not pursued by the clinical team. The patient underwent radiation without further imaging.

• 3. Nodular intradural lesions in the lumbar spine for which contrast-enhanced MRI was recommended for further evaluation and completed. These lesions enhanced and were concerning for recurrent ependymoma.

• 4. Infiltrative mass arising from the sacrum at the edge of the field of view and multiple marrow-replacing lesions throughout the spine. Contrast-enhanced MRI was recommended and completed. Work-up eventually revealed osteosarcoma of the sacrum with osseous metastatic disease.

• 5. Areas of heterogeneous marrow signal abnormalities for which contrast-enhanced MRI was recommended. The follow-up contrast-enhanced lumbar spine MRI was interpreted as most consistent with radiation-related marrow changes given history of prostate cancer, without concern for osseous metastatic disease. No change in management occurred.

• 6. Areas of marrow signal abnormality centered at the endplate, concerning for possible infection. Contrast-enhanced MRI was completed and findings were favored to be degenerative and the patient was discharged.

Discussion

The purpose of this study was to investigate the utility of an abbreviated total spine protocol for triage of emergency department patients presenting with clinical concern for acute spinal cord compression. In the retrospective phase of this study, the potential of an abbreviated protocol comprising of sagittal and axial T2w sequences through the entire spine was evaluated through retrospective review by readers blinded to history and additional sequences. While the positive predictive value varied by reader, the negative predictive value for ruling out spinal cord compression and/or severe spinal canal stenosis was 0.97 or greater for both readers. The sensitivity and specificity were greater than 78% for both readers. These results are comparable to the one previous study that retrospectively reviewed a similar protocol of sagittal STIR and axial T2w images. ⁶ These results suggest an abbreviated protocol may be an effective screening tool for ruling out cord compression in ER patients to facilitate triage and discharge.

The positive predictive value varied by reader and was lower than the negative predictive value. A possible contributing factor could have been the experience level of the blinded readers (5 and 6 years of professional experience) compared to the experience levels of the radiologists performing consensus evaluation (greater than 20 years), which served as the ground truth. Additionally, consensus evaluation for ground truth was performed with access to the specific clinical indication and prior imaging when applicable while the review performed by the readers was blinded. Blinded evaluation without clinical context or prior imaging may also have contributed to the moderate agreement in assessment of presence of cord compression.

The second aim of this study was to provide initial results from our institution following prospective implementation of an abbreviated cord compression protocol. To the authors’ knowledge, this study is the first in the literature reporting clinical findings prospectively following implementation of an abbreviated total spine protocol. The institutional change in work-flow occurred after consensus discussions with the neurosurgery and emergency medicine departments at our institution. While the initial protocol trialed during retrospective evaluation included only sagittal T2 and axial T2w images, the final protocol that was implemented included sagittal T1w images for better evaluation of marrow signal abnormalities and the sagittal T2 weighted sequence was replaced by sagittal T2 Dixon to allow for fat-suppressed images and therefore better evaluation of soft tissue edema. While sagittal T1w images do not contribute to evaluation of spinal cord compression, neuroradiologists in our division felt that it should be included for screening of clinically relevant marrow-replacing processes (i.e., osseous metastatic disease or osteomyelitis). Despite the addition and change in the sequences, the final cord compression protocol still yielded an approximately 60% reduction in imaging time on all 1.5T scanners compared to the total spine protocol without and with contrast, which was previously standard in our institution for emergency department patients presenting with clinical concern for cord compression. This imaging time is comparable to the previously reported study evaluating an abbreviated total spine protocol comprised of axial T2w images and sagittal STIR images. In that study, the authors reported a reduction in imaging time to approximately 24 min (on GE and Siemens 1.5T systems). ¹² This is comparable in length to the imaging time we have reported on our institution’s Siemens and Phillips 1.5T systems.

Initial results of the implementation have demonstrated an overall low call-back rate of 7% by the interpreting radiologist. Of the 6 cases where call-back was recommended, contrast-enhanced MRI was ordered by the clinical team in 5 of those cases and it only clarified or changed management in two of those cases. These results highlight that abbreviated non-contrast examination may be sufficient for screening evaluation in most emergency department patients presenting with concern for acute spinal cord compression.

Of the cases where concern was raised for cord compression, nerve root compression and/or severe spinal canal stenosis by the interpreting radiologist, operative management was undertaken in approximately 43% of cases. In cases where non-operative management was followed, the most common reasons were the patient left against medical advice or refused surgery, the level of spinal canal stenosis was not felt to explain the patient’s clinical presentation, or the patient’s presentation was not felt to be acute and operative management was scheduled as an outpatient after discharge. Of those patients who were taken to the operative room urgently during the same admission, dedicated further MRI imaging was acquired in 2/11 cases. Therefore, in most of these cases, the MRI images from the abbreviated protocol proved sufficient not only for the consultant neurosurgeon or orthopedic surgeon to make the decision to pursue operative management, but also proved sufficient for pre-operative planning. Of note, 3/11 patients did undergo pre-operative CT imaging as additional pre-operative work-up.

A small subset of the patients imaged in the prospective phase of this study underwent imaging on the Siemens 0.55 T Free.Max scanner (n = 8 of 85). Inclusion of the Siemens 0.55T system in imaging emergency department patients was reflective of acquisition of the scanner into the institutional fleet of systems between the retrospective and prospective phases of the study. Inclusion of the modern mid-field system introduced a source of variation and is a limitation of the study. However, only 1 of the 6 call-back cases were from patients imaged on this scanner. Moreover, despite expected diminished signal to noise ratio associated with the lower main magnetic field strength, previous work performed at our institution found a high inter-reader agreement on the imaging finding/diagnosis in lumbar spinal evaluation at 0.55T, ¹³ suggesting its feasibility for clinical spinal imaging.

This study has certain additional limitations. Firstly, the focus of this study was for effective triage of patients with acute cord compression. Patients where there is a concern for infection or in patients with known primary malignancy where there may be concern for leptomeningeal carcinomatosis should undergo total spine imaging with contrast-enhancement, and the abbreviated MRI protocol may not be sufficient for clinical disposition in these instances. Secondly, the abbreviated protocol that was trialed during the retrospective phase of this project is not identical to the protocol that was implemented. These changes were made after feedback was acquired from the division at large. However, our results show that even though the trialed protocol during retrospective review was shorter, two independent readers both showed a high negative predictive value with comparable call-back rates. Additionally, the major addition of sagittal T1w images to the protocol does not contribute to assessment of cord compression/severe spinal canal stenosis, the focus of the retrospective phase of the study. Thirdly, the focus of the prospective arm following our institution’s implementation of the new protocol was to provide initial experience with patient call-back and patient outcomes. Cases which were reported as negative by the interpreting radiologist were not re-reviewed, precluding any evaluation for false negatives. Finally, the present findings represent pilot results after the first 3 months of implementation. Continued analysis will be required to see the performance of this protocol after longer term implementation.

In summary, our results show that an abbreviated non-contrast total spine protocol, demonstrated through retrospective review, has potential for high negative predictive value in ruling out cord compression and severe spinal canal stenosis in the emergency department setting. This has potential for greatly reducing imaging time and therefore wait-time for MRI in the emergency department and reduce delays in patient triage and discharge. Our initial findings upon implementing a rapid cord compression protocol at our institution resulted in approximate 60% reduction in imaging time with a low call-back rate of 7%. Moreover, in patients who had cord compression and/or severe spinal canal stenosis, the acquired MRI images proved sufficient for both triage and treatment planning purposes in most patients.

ORCID iD

Shruti Mishra https://orcid.org/0000-0002-7384-009X