Abstract
Background:
Spine MRI is increasingly considered for patients presenting through the emergency department (ED). It was hypothesized that more MRIs are being obtained of the entire spine (relative to localized regions) over time, and there are inconsistencies in this practice.
Methods:
Data were abstracted from the 2010 to 2021 M151Ortho PearlDiver national, multi-insurance, administrative data set. Patients who received spine MRI within 7 days of presenting through the ED were categorized as having had localized MRI (cervical, thoracic, lumbar, cervical and thoracic, or thoracic and lumbar) or total spine (cervical, thoracic, and lumbar). Patient characteristics were compared between these categories with multivariable analyses. Furthermore, the rate of follow-up MRI within 14 days of initial MRI was assessed and compared.
Results:
Of the 275,999 patients identified as undergoing spine MRI, 93.25% were localized MRI and 6.75% were total spine MRI. Over a decade, total spine MRIs increased from 4.85% in 2010 to 12.38% in 2021 (P < 0.0001). Independent predictive factors for receiving a total spine MRI included indication (trauma, infection, neoplasm with odds ratio [OR], 1.30 relative to degenerative), patient factors (younger age [OR, 1.47 per decade], male sex [OR, 1.21], higher comorbidity [OR, 1.38]), and nonclinical factors (region of the country West [OR, 1.32 relative to South] and insurance plan (Medicaid OR, 1.23 and Medicare OR, 1.10 relative to commercial; P < 0.0001 for each except 0.0005 for Medicare). Follow-up MRIs within 14 days were more likely for total spine MRI than for the localized MRI group (7.28% vs. 5.26%, P < 0.0001).
Discussion:
For patients presenting through the ED, total spine MRIs represent a small but growing minority of spine MRI scans obtained. The nonclinical factors associated with this decision and mildly increased need for follow-up scans after total relative to localized MRI suggest room for greater consistency of practice.
MRI has become a routine part of spine care, especially for degenerative, trauma, infection, and/or neoplastic pathologies presenting to the emergency department (ED).1,2 Such scans may be of localized regions of the spine (cervical, thoracic, lumbar, cervical and thoracic, or thoracic and lumbar) or of the total spine (cervical, thoracic, and lumbar). Trends and factors associated with such decisions are of clinical interest.
There may be advantages to obtaining an MRI of the total spine. For infections and tumors, this can be advantageous to assess for multifocal aspects of pathology.3-5 For trauma, this can be advantageous to assess for fracture/ligamentous injury throughout the spine, especially for the obtunded or distracted patient.5 For degenerative indications, this can be advantageous if the etiology of symptoms are not clear, and there are red flags such as bowel/bladder dysfunction with concern of spine etiology.5,6 In each of these situations, pathology could be missed if MRI was focused on localized regions of the spine.
On the contrary, there can be disadvantages to obtaining an MRI of the total spine, most notably related to study duration and image resolve.7 One study looked at the utility of whole spine MRI following a whole spine-CT for patients who experienced blunt trauma and found that it was unlikely, in most cases, to alter management and advocated for targeted spine MRIs.8 In attempt to overcome the resource utilization of total spine MRIs, some have worked to evolve shorter sets of imaging sequences.7 Unfortunately, this approach may result in the need to reimage the spine if resolve/quality is inadequate. Not only may this contribute to additional patient discomfort secondary to undergoing an additional scan but also creates excess overall resource utilization.
Several studies have looked at the utility of total spine MRIs ordered in the ED. Morris M. et al. assessed adult patients who underwent a total spine MRI and found them to be of greatest utility for patients presenting with trauma, motor deficits, and bowel/bladder issues.9 In addition, Karmazyn et al10 compared cervical spine MRIs with total spine MRIs in children who experienced abusive head trauma and found that total spine MRIs identified some thoracolumbar injuries that otherwise could have been missed.
This study aimed to assess a large, national, multi-insurance population undergoing localized versus total spine MRIs for patients within 7 days presenting to an ED. The goal was to delineate trends, factors associated with, and need for follow-up scans related to such practices.
Methods
Study Cohort
This study used the 2010 through 2021 M151Ortho PearlDiver Mariner Patient Claims Database (PearlDiver Technologies). This data set is well-established for spine-related studies.11-16 As this database outputs aggregated and deidentified health information, our institutional review board has found studies based on this data set exempt from review.
Patients who underwent a spine MRI within 7 days of presenting to the ED were identified using Current Procedural Terminology (CPT) codes. Such MRIs were categorized as localized (cervical, thoracic, lumbar, cervical and thoracic, or thoracic and lumbar) or total spine (cervical, thoracic, and lumbar).
The following CPT codes identify location of spinal MRI: CPT-72141(MRI of the cervical spine without contrast), CPT-72142 (MRI of the cervical spine with contrast), CPT-72156 (MRI of the cervical spine both without and with contrast), CPT-72148 (MRI of the lumbar spine without contrast), CPT-72149 (MRI of the lumbar spine with contrast), CPT-72158 (MRI of the lumbar spine both without and with contrast), CPT-72146 (MRI of the thoracic spine without contrast), CPT-72147 (MRI of the thoracic spine with contrast), and CPT-72157 (MRI of the thoracic spine both without and with contrast). Total spine MRI included codes from each of the spinal regions.
Patient Characteristics/Follow-up MRI
Patient characteristics were then extracted from the data set. These include age, sex, Elixhauser Comorbidity Index (ECI, an aggregated summary of patient comorbidities11-14,17-20), insurance plan (commercial, Medicaid, and Medicare), region of the country according to U.S. Census Bureau definitions (Midwest, Northeast, West, and South), and diagnosis (degenerative/other or trauma/infection/neoplasm).
Follow-up MRI scans for this study included patients who had a subsequent spine MRI within 14 days of having had either a total spine MRI or localized spine MRI when they presented to the ED. The same CPT codes were used to identify spine MRI studies.
Data Analysis
Total and localized spine MRI utilization within 7 days for patients who presented to the ED were traced over the decade of the study. A two mean T-test compared the difference in the utilization of total and localized spine MRI from 2010 to 2021.
Characteristics of those undergoing a total spine MRI versus localized spine MRI were then tabulated and compared. Univariable comparisons of those who received a total spine MRI and those who received a localized spine MRI were done with Pearson chi squared test for categorical variables (sex, region, insurance plan, and diagnosis) and Welch T-test for continuous variables (age and ECI). Multivariable logistical regression was then done to assess independent predictors of having had a total spine MRI.
All statistical analyses were provided by the PearlDiver data set with statistical significance set at P < 0.05 for all tests. Figures were created using Microsoft Excel (Microsoft) and Prism (GraphPad Softwares).
Results
Total Versus Localized Spine MRI Utilization Over a Decade
A total of 275,999 spine MRIs were identified, of which total spine MRI was used for 18,632 (6.75%) compared with localized spine MRI for 257,367 (93.25%; Figure 1). The precent of total spine MRIs gradually increased over the years of the study from 4.85% in 2010 to 12.38% in 2021 (P < 0.0001; Figure 2).
Figure 1.
Pie chartshowingbreakdown of the patients who had a spine MRI within 7 days of presenting to the ED. The pie graph is shown with the total number of localized spine, 257,367 (93.25%), and total spine, 18,632 (6.75%), MRIs taken for patients who presented to the ED. ED = emergency department
Figure 2.
Bar graph showingthe percent of patients who received a total spine MRI traced over a decade. Percent of patients receiving a total spine MRI are compared with those who received a localized MRI. Each percentage of total spine MRIs are listed above their respective shaded region and compared with the standardized patients who received a localized MRI. Alpha level was set at 0.05, and comparison of 2010 total spine MRI use compared with 2021 total spine MRI use resulted in a P value of <0.0001.
Characterization of Those Who Underwent Total Relative to Localized MRI
Characteristics of patients who underwent total versus localized spine MRI are shown and compared with univariable analyses in Table 1. Those who had a total spine MRI were younger (average age, 45.69 vs. 54.86 years), more likely to be male sex (49.39% vs. 44.13%), had higher comorbidity burden (average ECI, 3.47 vs. 2.91), were of different geographic locations, insurance type, and were slightly more likely to be done for degenerative/other indications than trauma, infection, neoplasm indications (77.01% vs. 76.45%; P < 0.0001 for each).
Table 1.
Univariable Factors for Patients Admitted to the Emergency Department Who Received Either a Total Spine MRI or a Localized Spinal MRI
| Factor or Variable | Localized MRI | Total MRI | P Value |
| N = 275,999 | 257,367 (93.25%) | 18,632 (6.75%) | |
| Age (mean ± SD), yr | 54.86 ± 17.01 | 45.69 ± 23.45 | |
| <40 | 49,528 (19.24%) | 6695 (35.93%) | <0.0001 |
| 40–59 | 94,157 (36.58%) | 5416 (29.07%) | |
| 60–79 | 110,279 (42.85%) | 6332 (33.98%) | |
| >80 | 4851 (1.88%) | 287 (1.54%) | |
| Sex | |||
| Male | 113,572 (44.13%) | 9203 (49.39%) | |
| Female | 143,794 (55.87%) | 9429 (50.60%) | <0.0001 |
| ECI (mean ± SD) | 2.91 ± 2.70 | 3.47 ± 2.92 | |
| 0–1 | 95,814 (37.23%) | 5542 (29.74%) | |
| 2–3 | 76,529 (29.74%) | 5416 (29.07%) | <0.0001 |
| 4–6 | 57,696 (22.42%) | 4870 (26.14%) | |
| >6 | 27,309 (10.61%) | 2801 (15.03%) | |
| Insurance | |||
| Commercial | 214,560 (83.37%) | 15,465 (83.00%) | |
| Medicare | 47,728 (18.54%) | 2497 (13.40%) | |
| Medicaid | 22,341 (8.68%) | 2734 (14.67%) | <0.0001 |
| Region | |||
| South | 95,300 (37.03%) | 6377 (34.23%) | |
| Midwest | 69,478 (27.00%) | 4823 (25.89%) | |
| Northeast | 55,469 (21.55%) | 4004 (21.49%) | <0.0001 |
| West | 35,444 (13.77%) | 3186 (17.10%) | |
| Diagnosis | |||
| Degenerative/other | 196,762 (76.45%) | 14,360 (77.07%) | <0.0001 |
| Trauma, infection, neoplasm | 60,605 (23.55%) | 4272 (22.93%) |
ECI = Elixhauser Comorbidity Index
Multivariable analysis of patient characteristics of those who underwent total relative to localized spine MRI shown in Table 2 and Figure 3. Independent predictors of having had a total spine MRI in order of decreasing odds included: younger age (odds ratio [OR], 1.47 over decade increase), higher ECI (OR, 1.38 per two-point increase), region of the country (relative to South, West OR, 1.32), diagnosis (relative to degenerative/other, trauma, infection, or neoplasm OR, 1.30), insurance (relative to Commercial, Medicare OR, 1.10 [P value 0.0005] and Medicaid OR, 1.23), and sex (relative to female, male OR, 1.21; P < 0.0001 for each unless otherwise specified).
Table 2.
Multivariate Analysis of Predictive Factors for Receiving a Total Spine MRI for Patients Admitted to the Emergency Department
| N = 275,999 | OR (95% CI) | P Value |
| Age (per decade decrease) | 1.47 (1.45-1.47) | <0.0001 |
| Sex | ||
| Female (referent) | ||
| Male | 1.21 (1.17-1.25) | <0.0001 |
| ECI (per 2-point increase) | 1.38 (1.36-1.40) | <0.0001 |
| Region | ||
| South (referent) | ||
| Midwest | 1.02 (0.98-1.06) | 0.4509 |
| Northeast | 1.03 (0.99-1.07) | 0.1806 |
| West | 1.32 (1.26-1.39) | <0.0001 |
| Insurance | ||
| Commercial (referent) | ||
| Medicare | 1.10 (1.04-1.16) | 0.0005 |
| Medicaid | 1.23 (1.17-1.29) | <0.0001 |
| Diagnosis | ||
| Degenerative/other (referent) | ||
| Trauma, infection, or neoplasm | 1.30 (1.25-1.35) | <0.0001 |
CI = confidence interval, ECI = Elixhauser Comorbidity Index, OR = odds ratio
Significance for bold entries indicates 0.05 alpha level.
Figure 3.
Forest plot displaying independent risk factors for receiving a total spine MRI within 7 days of presenting to the ED. ED = emergency department
Patients Who Required an Additional Spine MRI Within 14 Days
Of the 275,999 spine MRI scans taken, additional spine MRI within 14 days was taken for 15,415 (5.59%). Of the 18,632 patients who underwent a total spine MRI, additional MRI was taken within 14 days for 1357 (7.28%). Of the 257,367 patients who underwent a localized spine MRI, additional MRI was taken withing 14 days for 14,058 (5.46%; Figure 4). These rates were significantly different (P < 0.0001).
Figure 4.
Bar graphs displaying the percentage of patients who underwent an additional spine MRI within 14 days of either having a total spine MRI or a localized spine MRI. Percentage of an additional MRI is listed above the dark shaded region, which represents those who had another scan. The percentages were compared at the 0.05 alpha level with a resulting P value of <0.0001.
Discussion
Spine MRI may be considered for patients presenting to the ED. Such MRIs can be localized to defined regions of the spine (typically higher resolve and/or shorter studies) or be of the total spine (typically lower resolve and/or longer studies). The trends in relative utilization of these two MRI modalities and drivers of utilization of localized versus total spinal MRI are not well characterized. With the increased attention on expenditures for medical imaging,21 understanding these trends and drivers are important. This study assessed localized versus total spine MRI utilization for patients who received a spinal MRI within 7 days of presenting to the ED tracked from 2010 to October 2021.
Over the entire study interval, the greatest fraction of patients presenting to the ED who received a spinal MRI received a localized MRI (93.25%, n = 257,367) relative to a total spinal MRI (6.75%, n = 18,632). However, the fraction of patients presenting to the ED receiving a total spinal MRI increased over the duration of the study (4.85% in 2010 vs. 12.38% in 2021, P < 0.0001). This trend suggests that practice patterns are evolving, with total spinal MRI being increasingly considered.
Several clinical factors were identified on multivariable analysis to be associated with total spine MRI. These include younger patient age, male sex, and lower comorbidity burden. This may be related to a higher index of suspicion for high-energy traumatic mechanisms among such patient populations,22 potentially leading to higher index of suspicion for multilevel pathology. To that end, those with nondegenerative diagnoses (trauma, infection, or neoplasm) were more likely to have total MRIs taken.
In addition, nonclinical factors were found to be independently associated with utilization of total spine MRI among patients presenting to the ED, such as geographic region and insurance. These findings are consistent with prior literature demonstrating variability in spine-related imaging and interventions based on regional practice patterns, payer type, and resource availability.23-25 Perceived malpractice risk has also been associated with increased imaging use in emergency settings,26 raising the possibility that defensive medicine may contribute to broader imaging strategies such as total spine MRI. The association of total spine MRIs suggest the lack of consensus and nonclinical divers of such decision making that could be addressed in striving for best practices.
Although limited data exist specifically on total spine MRI utilization trends, the current findings reflect patterns observed in broader spine imaging literature, where increasing imaging rates have been reported, particularly in urban or higher-resource settings.27 The current findings contribute to the growing body of evidence on imaging utilization and emphasize the need to examine institutional and systemic factors that shape practice patterns.
With advantages and disadvantages of larger field total spine MRI, there is the question of need for repeat MRI following localized or total spine MRI. This was assessed in the 2 weeks following initial MRI was assessed. Patients who received a total spine MRI initially had higher rates of rescan within this period compared with those who initially underwent localized spine MRI (7.28% vs. 5.46%, P < 0.0001), perhaps due to the relatively lower resolve of the larger scan areas of the total spinal MRI and need for follow-up images at spinal levels of concern. These findings raise questions about the diagnostic efficiency and downstream clinical effect of broader versus more targeted imaging approaches.
This study had several limitations, including those inherent to retrospective studies, such as potential coding errors in the electronic health record, and inability to analyze variables not captured in claims/the electronic health record. Furthermore, it was not possible to determine the specific clinical reasoning for obtaining the imaging that was done.
For patients presenting through the ED, total spine MRIs represent a small but growing minority of spine MRI scans being obtained. The nonclinical factors associated with this decision and mildly increased need for follow-up scans after total relative to localized MRI suggest room for greater consistency of practice.
Footnotes
Dr. Grauer or an immediate family member serves as Editor-in-Chief of the North American Spine Society Journal (NASSJ). None of the following authors or any immediate family member has received anything of value from or has stock or stock options held in a commercial company or institution related directly or indirectly to the subject of this article: Ms. Rudisill, Dr. Dhodapkar, Mr. Ratnasamy, and Dr. Jeong.
Contributor Information
Katelyn E. Rudisill, Email: katelyn.rudisill@temple.edu.
Meera M. Dhodapkar, Email: meera.dhodapkar@yale.edu.
Philip P. Ratnasamy, Email: philip.ratnasamy@yale.edu.
Seongho Jeong, Email: seongho.jeong@yale.edu.
References
- 1.Winn A, Martin A, Castellon I, et al. : Spine MRI: A review of commonly encountered emergent conditions. Top Magn Reson Imaging 2020;29:291-320. [DOI] [PubMed] [Google Scholar]
- 2.Kumar Y, Hayashi D: Role of magnetic resonance imaging in acute spinal trauma: A pictorial review. BMC Musculoskelet Disord 2016;17:310. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Nakanishi K, Tanaka J, Nakaya Y, et al. : Whole-body MRI: Detecting bone metastases from prostate cancer. Jpn J Radiol 2022;40:229-244. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Balogova S, Zakoun JB, Michaud L, et al. : Whole-body 18F-fluorocholine (FCH) PET/CT and MRI of the spine for monitoring patients with castration-resistant prostate cancer metastatic to bone: A pilot study. Clin Nucl Med 2014;39:951-959. [DOI] [PubMed] [Google Scholar]
- 5.Sergi CM, Miller E, Demellawy DE, Shen F, Zhang M: Chronic recurrent multifocal osteomyelitis. A narrative and pictorial review. Front Immunol 2022;13:959575. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Ha JH, Lee JH, Lee JH: Coexisting spine lesions on whole spine T2 sagittal MRI in evaluating spinal degenerative disease. J Korean Med Sci 2021;36:e48. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Oztek MA, Brunnquell CL, Hoff MN, et al. : Practical considerations for radiologists in implementing a patient-friendly MRI experience. Top Magn Reson Imaging 2020;29:181-186. [DOI] [PubMed] [Google Scholar]
- 8.Atsina KB, Rozenberg A, Selvarajan SK: The utility of whole spine survey MRI in blunt trauma patients sustaining single level or contiguous spinal fractures. Emerg Radiol 2019;26:493-500. [DOI] [PubMed] [Google Scholar]
- 9.Morris M, Destian S, Chu Y, Klumpp M, Zohrabian VM: Investigation of whole spine MRI in the emergency department at two large tertiary care academic medical centers in the United States. Curr Probl Diagn Radiol 2021;50:637-645. [DOI] [PubMed] [Google Scholar]
- 10.Karmazyn B, Reher TA, Supakul N, et al. : Whole-spine MRI in children with suspected abusive head trauma. AJR Am J Roentgenol 2022;218:1074-1087. [DOI] [PubMed] [Google Scholar]
- 11.Rudisill KE, Ratnasamy PP, Maloy GC, Grauer JN: Decline in separate incision autograft for spine surgery over the past decade: A fading “gold standard”. J Am Acad Orthop Surg 2023;31:938-944. [DOI] [PubMed] [Google Scholar]
- 12.Rudisill KE, Ratnasamy PP, Joo PY, Rubin LE, Grauer JN: Magnetic resonance imaging in the year prior to total knee arthroplasty: A potential overutilization of healthcare resources. J Am Acad Orthop Surg Glob Res Rev 2023;7:e22.00262. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Ratnasamy PP, Rudisill KE, Oghenesume OP, Riedel MD, Grauer JN: Risk of contralateral hip fracture following initial hip fracture among geriatric fragility fracture patients. J Am Acad Orthop Surg Glob Res Rev 2023;7:e23.00001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Ratnasamy PP, Rudisill KE, Joo PY, Lattanza LL, Grauer JN: Trends in open versus endoscopic carpal tunnel release from 2010 to 2021. J Am Acad Orthop Surg Glob Res Rev 2024;8:e24.00077. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Gouzoulis MJ, Kammien AJ, Zhu JR, Gillinov SM, Moore HG, Grauer JN: Single-level posterior lumbar fusions in patients with Ehlers Danlos Syndrome not found to be associated with increased postoperative adverse events or five-year reoperations. N Am Spine Soc J 2022;11:100136. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Ratnasamy PP, Maloy GC, Oghenesume OP, Peden SC, Grauer JN, Oh I: The burden of revision total ankle replacement has increased from 2010 to 2020. Foot Ankle Orthop 2023;8:24730114231198234. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Rudisill KE, Ratnasamy PP, Sanchez JG, Grauer JN: Disparities in dual-energy X-ray absorptiometry scan utilization across race/ethnic groups before and after hip fractures. J Am Acad Orthop Surg Glob Res Rev 2024;8:e24.00052. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Day W, Gouzoulis MJ, Jayaram RH, Grauer JN: Emergency department utilization after elbow arthroscopy. JSES Int 2024;8:910-914. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Dhodapkar MM, Modrak M, Halperin SJ, Joo P, Luo X, Grauer JN: Trends in and factors associated with surgical management for closed clavicle fractures. J Am Acad Orthop Surg Glob Res Rev 2023;7:e23.00226. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Dhodapkar MM, Gouzoulis MJ, Halperin SJ, Modrak M, Yoo BJ, Grauer JN: Urgent care versus emergency department utilization for foot and ankle fractures. J Am Acad Orthop Surg 2023;31:984-989. [DOI] [PubMed] [Google Scholar]
- 21.Iglehart JK: Health insurers and medical-imaging policy—A work in progress. New Engl J Med 2009;360:1030-1037. [DOI] [PubMed] [Google Scholar]
- 22.Ziegenhain F, Scherer J, Kalbas Y, et al. : Age-dependent patient and trauma characteristics and hospital resource requirements-can improvement be made? An analysis from the German trauma registry. Medicina (Kaunas) 2021;57:330. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Deyo RA, Mirza SK, Turner JA, Martin BI: Overtreating chronic back pain: Time to back off? J Am Board Fam Med 2009;22:62-68. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Raja AS, Ip IK, Sodickson AD, et al. : Radiology utilization in the emergency department: Trends of the past 2 decades. AJR Am J Roentgenol 2014;203:355-360. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Wennberg JE, Fisher ES, Goodman DC, Skinner JS: Tracking the Care of Patients With Severe Chronic Illness: The Dartmouth Atlas of Health Care 2008. Lebanon, NH: The Dartmouth Institute for Health Policy and Clinical Practice, 2008. https://www.ncbi.nlm.nih.gov/books/NBK586192/ [PubMed] [Google Scholar]
- 26.Studdert DM, Mello MM, Sage WM, et al. : Defensive medicine among high-risk specialist physicians in a volatile malpractice environment. JAMA 2005;293:2609-2617. [DOI] [PubMed] [Google Scholar]
- 27.Baker LC, Atlas SW, Afendulis CC: Expanded use of imaging technology and the challenge of measuring value. Health Aff (Millwood) 2008;27:1467-1478. [DOI] [PubMed] [Google Scholar]