Enhancing back pain and sciatica diagnosis: Coronal short tau inversion recovery’s role in routine lumbar magnetic resonance imaging protocols

Abstract

BACKGROUND

Back pain and sciatica are common complaints that often require imaging for accurate diagnosis and management. Conventional lumbar magnetic resonance imaging (MRI) protocols typically include sagittal and axial T1 and T2 sequences; however, these may miss certain pathologies. The addition of coronal short tau inversion recovery (STIR) sequences offers the potential to enhance the detection of both spinal and extra-spinal abnormalities, thereby improving clinical decision-making and patient outcomes.

AIM

To evaluate the impact of adding coronal STIR sequences to routine lumbar MRI in diagnosing back pain and sciatica.

METHODS

We prospectively analyzed data from patients aged 6 and older presenting with back pain or sciatica who underwent lumbar spine MRI at our institution. The standardized MRI protocol utilized included sagittal and axial T1 and T2 sequences, complemented by a coronal STIR sequence. Data on structural abnormalities were collected, reviewed, and analyzed using counts, percentages, and Fisher's exact test for categorical variables.

RESULTS

Our cohort comprised 274 patients (115 males, 159 females; mean age 44.91 years). Notably, 39 patients exhibited abnormalities across all sequences, while 72.63% showed normal findings on the coronal STIR sequence. Importantly, 30.29% of cases were diagnosed as normal without the coronal STIR, and 36 patients with normal T1 and T2 sequences presented abnormalities on the coronal STIR. The coronal STIR sequence successfully identified 26 spinal and 10 non-spinal pathologies, including 17 cases of sacroiliitis, with a significant association (P < 0.0001) between sacroiliitis diagnosis and abnormalities visible solely on this sequence.

CONCLUSION

Integrating coronal STIR into routine lumbar MRI enhances detection of hidden spinal and extra-spinal pathologies, improves patient management, and offers a cost-effective, practical upgrade with significant diagnostic and clinical value.

Key Words: Low back pain; Magnetic resonance imaging; Coronal short tau inversion recovery; Sacroiliitis; Diagnostic accuracy

Core Tip: Incorporating a coronal short tau inversion recovery sequence into routine lumbar magnetic resonance imaging protocols significantly enhances diagnostic accuracy for patients with back pain and sciatica. This sequence improves the detection of spinal and extra-spinal pathologies - especially sacroiliitis and pelvic abnormalities - that may be missed on standard axial and sagittal T1 and T2 sequences. Our findings support the integration of coronal short tau inversion recovery as a cost-effective, practical addition that can meaningfully influence diagnosis and clinical management.

INTRODUCTION

Low back pain is linked to considerable morbidity, contributing to around 850 years lived with disability in Jordan in 2021, according to data from the Global Burden of Disease Study, 2021[1]. Back pain is one of the most common reasons for performing routine lumbar magnetic resonance imaging (MRI), which include T1-weighted, T2-weighted, and fat-suppressed fluid-sensitive sequences in both the sagittal and axial planes[2-4]. The protocol can also include transverse plane images focused on intervertebral levels with structural abnormalities identified in the sagittal view. Although sagittal short tau inversion recovery (STIR) imaging is widely employed for evaluating vertebral marrow abnormalities and spinal alignment, its relatively narrow field of view (FOV) often excludes important extraspinal structures. In contrast, coronal STIR sequences allow for a broader anatomical assessment, encompassing the sacroiliac joints, hips, and pelvic organs, thereby increasing the likelihood of detecting clinically relevant findings beyond the spine. Romeo et al[5] demonstrated that coronal STIR imaging revealed additional extraspinal abnormalities in 6.4% of patients undergoing lumbar MRI, with 72% of these findings visible only in the coronal plane, underscoring the complementary role of this sequence in routine imaging protocols and its potential to impact clinical decision-making[6].

This standard protocol effectively identifies common causes of back pain and sciatica, such as disc prolapse, yet may miss extra-spinal causes. Incorporating a coronal STIR sequence can enhance the detection of conditions outside the spine, such as sacroiliac, hip joint, and abdominopelvic pathologies[7]. The STIR sequence in MRI is designed to suppress fat signals, enhancing the visibility of certain pathologies like edema or inflammation[8]. Based on findings from previous studies and our experience, we began incorporating coronal STIR sequences into our lumbar spine MRI protocol for patients with back pain or sciatica, especially when standard sequences yield inconclusive results. This addition provides a broader view of the abdominal and pelvic regions, allowing assessment of sacroiliac, hip joints, and abdominopelvic organs, which may contribute to the patient’s symptoms. This study aims to evaluate the impact of adding coronal STIR imaging to routine lumbar MRI, focusing on its potential to uncover hidden causes and improve diagnostic accuracy for back pain and sciatica.

MATERIALS AND METHODS

Ethical considerations

This study followed ethical standards and received approval from the institutional review board. All patient data were handled in compliance with confidentiality regulations.

Study design

This study evaluated prospectively collected data from patients presenting with back pain and sciatica who underwent lumbar spine MRI. Patient medical records and imaging data were reviewed to identify findings relevant to the study’s aim.

Participant selection

The study included patients aged 6 or above who reported back pain and sciatica as primary complaints and underwent a lumbar spine MRI at our hospital to identify potential underlying pathologies.

MRI protocol

All participants underwent a standardized MRI protocol on a GE Signa HD 1.5T MRI System. The routine protocol included sagittal and axial T1- and T2-weighted sequences, with an additional coronal STIR sequence. These sequences provided a detailed evaluation of the spine and surrounding structures to detect abnormalities potentially related to back pain.

Lumbosacral MRI protocol

Lumbosacral MRI protocol including: (1) Sagittal T2-weighted imaging: Echo time (TE): 102 milliseconds, repetition time (TR): 2584 milliseconds, slice thickness: 4 mm, FOV: 30 cm; (2) Sagittal T1-weighted imaging: TE: Minute full, TR: 584 milliseconds, slice thickness: 4 mm, FOV: 30 cm; (3) Axial T2-weighted imaging: TE: 110 milliseconds, TR: 2600 milliseconds, slice thickness: 4 mm, FOV: 22 cm; and (4) Axial T1-weighted imaging: TE: Minute full, TR: 450 milliseconds, slice thickness: 4 mm, FOV: 22 cm.

Coronal STIR sequence

TE: 50 milliseconds, TR: 4050 milliseconds, slice thickness: 5 mm, FOV: 48 cm.

Data collection

Data were prospectively gathered from observations across the T1/T2 axial and coronal STIR sequences, with a focus on structural abnormalities and variations. An experienced radiologist reviewed the images to confirm these findings.

Statistical analysis

The collected findings from the MRI scans, encompassing abnormalities and structural variations observed in the T1/T2 axial sequences and the coronal STIR sequence, were summarized as counts and percentages. A Fisher’s exact test was performed to assess the significance of differences observed between categorical variables. Statistical analysis was conducted using Prism GraphPad software version 10.1.2.

RESULTS

Our sample included all the patients who underwent spine MRI for evaluation of low back pain at our hospital. 115 male and 159 female patients with a mean age of 44.91 years (range 6-92 years) were included. The most common findings on T1 and T2 sequences, as well as coronal STIR sequences, are evident in Table 1. 39 patients showed abnormalities in all T1, T2, and coronal STIR sequences. Most of the cases (72.63%) had normal findings on the coronal STIR sequence. Without using coronal STIR, MRI examination was considered within normal limits in 83 subjects (30.29%). Total of 36 patients who did not show abnormalities on T1 and T2 sequences showed abnormality on coronal STIR; 26 had spinal pathologies, and 10 were non-spinal. The pathologies of these 36 cases included sacroiliitis (17 cases), fractures (4), uterine fibroids (2), ovarian abnormalities (2), scoliosis (2), Legg-Calve-Perthes disease (1), fallopian tube abnormalities (1), fibroma in the erector spinae muscle (1), hematoma (1), metabolic bone disease (1), liver and pelvic mass (1), spondylitis (1), inflammatory bone process (1), and pilonidal sinus (1). STIR sequences had added merit in detecting non-spinal pathologies in 26 patients without being found on T1 and T2 sequences. Of the 17 cases of sacroiliitis, the average age was 37.1%, and the age range was between (18-81); the median age was 36. Of them, 11 were female, and 6 were male. 5 cases were bilateral sacroiliitis. 16 patients had findings suggestive of a neoplastic process on coronal STIR sequence, with only 1 of them being shown on T1 and T2 sequences. To investigate whether the number of cases diagnosed as sacroiliitis was significantly associated with abnormalities only found with the coronal STIR sequence, a Fisher’s exact test was run. shows that most cases of sacroiliitis showed pathogenic abnormalities on the coronal STIR sequence, with a significant P-value of < 0.0001.

Table 1 Pathologic findings on magnetic resonance sequences.

	Number of cases	Percentage
Pathology evident on T1 and T2 sequences
Disc prolapse	147	53.65%
Canal stenosis	39	14.23%
Spondylolisthesis	9	3.28%
Schmorl’s node (non-pathologic finding)	5	1.82%
Fracture	4	1.46%
Spinal dysraphism	2	0.73%
Findings suggestive of a neoplastic process	1	0.36%
Spondylodiscitis	1	0.36%
Pathologies outside the vertebral structures	0	0.00%
Non-pathologic findings (normal)	83	30.29%
Pathology evident on coronal STIR sequence
Pathologies outside the vertebral structures	26	9.49%
Sacroiliitis	23	8.39%
Findings suggestive of a neoplastic process	16	5.84%
Fracture	4	1.46%
Spinal dysraphism	1	0.36%
Legg-Calve-Perthes disease	1	0.36%
Non-pathologic findings (normal)	199	72.63%

STIR: Short tau inversion recovery.

DISCUSSION

Our study, utilizing prospectively collected data from patients evaluated for back pain and sciatica using T1, T2, axial, and coronal STIR sequences, demonstrates that incorporating coronal STIR imaging can significantly alter expected diagnoses and impact patient management. For instance, a study by Zeitoun and Mohieddin[9] found that adding coronal STIR sequences to routine lumbar MRI protocols detected extra-spinal abnormalities in 20% of patients, leading to changes in diagnosis and influencing management in 6.9% of cases.

Our findings align with previous research highlighting the diagnostic value of incorporating coronal STIR sequences into routine lumbar spine MRI protocols. Patriat et al[8], in a large cohort study of 600 Lumbar spine MRIs, demonstrated that the addition of a large-field coronal STIR sequence led to the detection of extra-spinal abnormalities in 11.3% of cases - many of which, including gluteal tendinobursitis, hip osteoarthritis, and sacroiliitis, were not visible on standard sagittal images. Importantly, the study also reported incidental findings with potential clinical relevance in 11.7% of patients, emphasizing the broader diagnostic reach of this sequence. Similarly, Romeo et al[5] assessed the impact of coronal STIR imaging in patients presenting with low back pain and found additional extraspinal findings in 6.4% of cases, with the majority (72%) detectable exclusively on the coronal plane. Notably, the incorporation of this sequence influenced clinical decision-making in 3.5% of cases. These findings support the growing consensus that coronal STIR imaging offers meaningful diagnostic and therapeutic benefits, particularly in detecting pathologies that may otherwise go unrecognized on conventional imaging planes[6,10].

The inclusion of coronal STIR sequences identified abnormalities in 30.29% of cases, despite normal T1 and T2 imaging. This underscores its importance in diagnosing both spinal and non-spinal pathologies, with 26 cases linked to spinal issues and 10 to non-spinal conditions. Notably, Fisher's exact test revealed a significant association between the diagnosis of sacroiliitis and abnormalities identified solely via the coronal STIR sequence (P < 0.0001).

In response to these findings, our institution has adopted coronal STIR imaging as a standard part of the lumbar MRI protocol, aiming to expedite diagnosis and improve treatment efficacy while potentially reducing unnecessary further investigations. However, this protocol change may have cost implications that necessitate additional cost-effectiveness studies.

A 2019 study in Egypt involving 50 patients corroborates our results regarding the benefits of coronal STIR sequences, identifying a 20% increase in diagnostic findings, while our study found an additional 13.14%[9]. Both studies highlight coronal STIR’s ability to unveil significant pathologies not discernible in routine MRI sequences, enhancing diagnostic accuracy. We excluded patients with a history of surgical follow-up for back imaging; however, traumatic patients were included. We found abnormalities in 85% of T1 and T2 sequences, with 14.23% of cases showing abnormalities across all sequences, and 13.14% detected exclusively on coronal STIR. Thus, the number of patients with abnormal findings increased by 13.14%.

A recent study involving 600 consecutive lumbar MRI scans, performed using the same protocol as our investigation, retrospectively reviewed cases with two musculoskeletal radiologists independently assessing coronal STIR sequences for extra-spinal abnormalities contributing to lumbar radiculopathy[9]. Extra-spinal cause on the coronal STIR sequence in 68 cases (11.3%), primarily gluteal tendon bursitis (30.9%), congestive hip osteoarthritis (25%), degenerative sacroiliac arthropathy (14.7%), and inflammatory sacroiliitis (7.3%)[8]. In our study, we observed additional abnormalities after incorporating coronal STIR in 36 patients who had normal T1 and T2 findings, with sacroiliitis being the most common finding at a rate of 47.2% (17 patients).

Extra spinal injuries were frequently noted, with 70 cases (11.7%), of which 38% involved tumors or pseudo-tumor masses necessitating further evaluation[8]. In our study, we identified extraspinal injuries in 13.8% of the 36 patients, including fractures and hematomas. A previous study indicated higher rates of definite extra-spinal sciatica causes in patients aged 24 to 40[7]. In our study of 274 patients, the mean ages for abnormal and normal coronal STIR were 43.96 and 45.26, respectively, with no significant age difference (P = 0.5604), Table 2. Of these, 199 had normal coronal STIR findings, but 36 showed abnormalities undetected by T1 and T2 sequences, with 26 having spinal and 10 non-spinal pathologies.

Table 2 Coronal short tau inversion recovery results and patients’ characteristics, n (%).

Characteristics	Abnormal coronal STIR	Normal coronal STIR	P value
Mean age	43.96	45.26	0.5604a
Females	47 (29.56)	112 (70.44)	0.4103b
Males	28 (24.35)	87 (75.65)

^aUnpaired t-test.

^bFisher’s exact test.

STIR: Short tau inversion recovery.

Comparing our results with a study of 931 patients, which found a 6.4% rate of additional extraspinal findings, we observed a higher percentage (13.14%). Our study also revealed that around 10.9% had changes in their treatment workflow[6]. A 2015 study with a similar demographic profile highlighted degenerative disc disease as the most common finding; however, our research identified a broader range of additional pathologies, including liver and pelvic abnormalities[10]. The differences in the FOV of coronal STIR images likely contributed to the higher abnormality rate in our study. This may be one of the reasons for the higher percentage of abnormal cases in our study. A finding such as femoral head avascular necrosis will not be shown in this study as the hip is not shown in the FOV Figure 1.

Figure 1 A 34-year-old male patient with low back and groin region pain. A: Sagittal T2-weighted imaging showing muscular spasm; B and C: Coronal short tau inversion recovery and weighted imaging showing bone marrow edema of the femoral head bilaterally suggestive of Legg-Calve-Perthes disease.

As mentioned, we have a percentage of 72% of the 36 patients with abnormal coronal STIR and normal T1 and T1, which is 9.5% of the total number of patients we included in our study, have a non-spinal pathology. The most evident were ureteric dilatation Figure 2, ovarian cysts, renal cysts, fallopian tube abnormalities, and masses, which change the treatment for the patient Figure 3. Sixteen patients had findings suggesting a neoplastic process on the coronal STIR sequence, with only one of these being detected on the T1 and T2 sequences (a bony lesion in the sacral area). The findings included seven ovarian cysts, four sacroiliac bony lesions, two uterine fibroids, one fibroma in the erector spinae muscle, one lung lesion, and one liver and pelvic mass. To determine if the number of sacroiliitis cases was significantly associated with abnormalities found solely on the coronal STIR sequence, a Fisher’s exact test was conducted, indicating that most cases of sacroiliitis exhibited pathogenic abnormalities exclusively on the coronal STIR sequence, with a significant P value of < 0.0001. This result shows that sacroiliitis is the most common finding in our study on the coronal STIR sequence Figure 4.

Figure 2 A 59-year-old male patient with low back pain. A: Sagittal weighted imaging showing disc protrusion at the level of L5-S1; B: Coronal short tau inversion recovery showing left ureteral dilatation, and the distal aspect (pelvic ureter) measuring about 10.05 mm.

Figure 3 Dual-patient magnetic resonance imaging findings: lumbar spine and pelvic pathologies. A: Patient A: Normal lumbar spine on sagittal weighted imaging; B: Patient A: Small ovarian cystic lesions on coronal short tau inversion recovery; C: Patient B: Normal lumbar spine on sagittal T2-weighted imaging; D: Patient B: Bilateral hydrosalpinx on coronal short tau inversion recovery.

Figure 4 Sagittal and coronal magnetic resonance imaging findings in a 38-year-old male with low back pain: Normal lumbar spine and sacroiliitis. A: Sagittal weighted imaging of a 38-year-old male with low back pain and right sciatica shows lumbar spine straightening without significant disc pathology; B: Coronal short tau inversion recovery reveals bone marrow edema, more on the right, indicating sacroiliitis.

Our study had several limitations that should be addressed in future research. Firstly, in cases of severe pain, we opted not to proceed with the examination or add the coronal STIR sequence. When abnormalities were detected on T1 and T2 images, the examination was halted to avoid discomfort for the patient. Claustrophobia also posed a potential barrier to including this additional sequence, though it is noteworthy that very few patients reported concerns about the extended imaging time. Moreover, while the incorporation of coronal STIR can reveal diseases that are difficult to diagnose with a single sequence, it may necessitate further imaging, which could lead to additional procedural requirements.

The inclusion of coronal STIR imaging revealed additional pathologies that may be difficult to diagnose with single-sequence evaluations, emphasizing the need for comprehensive imaging protocols. To enhance future research, we recommend integrating 3D imaging techniques into MRI protocols, such as coronal STIR 3D SPACE imaging. This method has demonstrated superior diagnostic capabilities by providing detailed anatomical depictions and multi-angular views, which can improve the assessment of complex conditions like spinal nerve pathologies.

CONCLUSION

This prospective study underscores the clinical importance and diagnostic value of incorporating coronal STIR sequences into routine lumbar MRI protocols for patients with back pain and sciatica. By significantly enhancing the detection of both spinal and extra-spinal pathologies many of which were missed on standard sequences, this approach represents a novel, cost-effective refinement to conventional imaging strategies, with direct implications for more accurate diagnosis, targeted treatment, and improved patient outcomes.

ACKNOWLEDGEMENTS

We sincerely thank the Radiology Team at KHCC for their invaluable assistance in data collection and analysis.