Important postoperative magnetic resonance imaging findings correlated with clinical outcomes to evaluate adequacy and success of decompression surgery–what radiologists need to know: a systematic review

Article information

Asian Spine J. 2025;.asj.2025.0061

Publication date (electronic) : 2025 September 22

doi : https://doi.org/10.31616/asj.2025.0061

Ifran Saleh¹, Rahmad Mulyadi², Witantra Dhamar Hutami¹, Kevin Dilian Suganda³

¹Department of Orthopaedic and Traumatology, Dr. Cipto Mangunkusumo National Central Public Hospital, Faculty of Medicine Universitas Indonesia, Jakarta, Indonesia

²Department of Radiology, Dr. Cipto Mangunkusumo National Central Public Hospital, Faculty of Medicine Universitas Indonesia, Jakarta, Indonesia

³Department of Radiology, Hasan Sadikin Central Public Hospital, Faculty of Medicine Universitas Padjadjaran, Bandung, Indonesia

Corresponding author: Rahmad Mulyadi, Department of Radiology, Dr. Cipto Mangunkusumo National Central Public Hospital, Faculty of Medicine Universitas Indonesia, Jalan Pangeran Diponegoro No. 71, Kenari, Senen, Kota Jakarta Pusat, Daerah Khusus Ibukota, Jakarta 10430, Indonesia, Tel: +62-8558801965, Fax: +62-21-3134 8991, E-mail: dr_rahmad_radiologi@yahoo.com

Received 2025 January 29; Revised 2025 June 14; Accepted 2025 June 16.

Abstract

Introduction

Degenerative lumbar spinal stenosis (LSS) is the most prevalent degenerative spinal disorder in older adults. The clinical signs include neurogenic claudication, low back pain, and lower extremity paresthesia, which can be refractory to conservative treatment. Refractory LSS patients are treated with decompression surgery, which has been shown to yield superior clinical outcomes compared to conservative treatment [1].

Current literature on postoperative radiologic findings, particularly magnetic resonance imaging (MRI), after decompression surgery is limited and inconsistent. Reviewing both normal and abnormal postoperative MRI findings is crucial, especially in patients with postoperative pain. To the authors’ knowledge, few studies have evaluated expected postoperative MRI findings and their correlation with clinical outcomes after decompression surgery. This study aimed to systematically review the best available evidence on MRI findings, particularly postoperative ones that correlate with the clinical outcomes, thereby assessing the adequacy and success of decompression surgery.

Materials and Methods

The authors of this systematic review adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) criteria [2]. This study aimed to evaluate the correlation between postoperative MRI findings and clinical outcomes in LSS patients who underwent decompression surgery. The systematic literature review commenced with the formulation of research questions guided by the PICO framework (patient/population/problem, comparison, intervention, and outcome). The research question is as follow: (1) Problem: inadequate spinal decompression; (2) Intervention: spinal decompression surgery; (3) Comparison: preoperative versus postoperative MRI findings; and (4) Outcome: clinical outcomes.

Following the formulation of research questions, the next step was to develop a library search strategy. A literature search was performed using keywords and index terms (including MeSH/Medical Subject Headings) across the Cochrane Library, Medline/PubMed, and EBSCOHost databases. The MeSH term used during literature search was “lumbar spinal stenosis” AND “postoperative magnetic resonance imaging.” Boolean operators (AND/OR) were utilized to optimize the search strategy. All aforementioned databases were searched up to June 2024. Additional relevant studies were identified through manual screening of the reference list.

The criteria for inclusion were as follows: (1) study design: randomized controlled trials (RCTs) or non-RCTs evaluating the correlation of preoperative and postoperative MRI findings and corresponding clinical outcomes; (2) participants: patients with degenerative LSS requiring decompression surgery with or without spinal instrumentation; (3) intervention: decompression surgery performed with or without instrumentation; and (4) outcomes evaluation: preoperative and postoperative MRI findings alongside corresponding clinical outcomes.

The criteria for exclusion were as follows: (1) cohort studies, review articles and meta-analysis, case reports or case series, and expert opinion; (2) animal or cadaveric studies; and (3) publications not written in English.

The methodological quality of each study obtained from the literature review was assessed using the level of evidence in medicine criteria outlined by Sackett [3]. The risk of bias in the included studies was assessed employing the criteria presented in the Cochrane Handbook. Assessment of quality and risk of bias in non-RCT studies was performed using the Newcastle-Ottawa Scale, as recommended by the Cochrane Collaboration [4].

Results

Literature search

Fig. 1 illustrates the literature search results. After applying the inclusion and exclusion criteria, we identified a total of 12 articles. The initial search yielded 394 articles, including 17 from PubMed/Medline, 29 from EBSCOHost, one from Cochrane Library databases, and 347 from Google Scholar. An additional four articles were identified through cross-references. Three hundred and sixteen articles were excluded based on their titles, 40 articles were excluded based on the abstract, and 17 articles were excluded after full-text review. Six articles were excluded due to duplication. After reviewing the cross-reference, four articles were included. The characteristics of the included studies are summarized in Table 1.

Fig. 1

Flow chart of literature search from electronic databases.

Table 1

Characteristic of the included studies

Methodological assessment

All the included studies were non-RCT. Quality evaluation of non-RCT studies was carried out using the Newcastle-Ottawa Scale. There were 12 non-RCT studies included in the systematic literature review. All non-RCT studies were good quality studies (Table 2). The characteristics of the included studies are presented in Table 1. Of 12 studies, two employed a prospective study design. One study did not specify the type of MRI machine used. Nine studies used a 1.5 Tesla MRI scanner, while the other two studies used 0.6 and 1.0 Tesla MRI machines. T2-weighted images were used in all MRI evaluations.

Table 2

Assessment of non-randomized controlled trials studies using Newcastle-Ottawa Scale

Magnetic resonance imaging findings

Dural sac cross-sectional area expansion

A decline in dural sac cross-sectional area (DSCA) is linked to degenerative LSS. On axial MRI, DSCA is measured as the narrowest area bounded by an imaginary line drawn around the area between the facet and lamina (Fig. 2). Posterior decompression, performed by a spine surgeon, is intended to alleviate the stenosis and relieve compression of the affected nerves. Successful decompression is typically associated with an expected expansion of the DSCA. Studies reveal that an absolute threshold of DSCA of <70 mm² after decompression surgery elevated the risk of persistent radiculopathy. In contrast, a DSCA of 70 mm² or more accompanied by symptom resolution can be considered indicative of adequate DSCA expansion and, consequently, adequate decompression.

Fig. 2

Measurement of dural sac cross sectional area on axial T2-weighted image magnetic resonance imaging.

Eight studies examined postoperative DSCA expansion following decompression surgery (Table 3) [5–12]. Only one study reported no correlation between postoperative DSCA expansion and clinical outcomes. The other seven studies demonstrated that significant postoperative DSCA expansion, compared to preoperative values, was correlated with improved clinical outcomes in terms of Oswestry Disability Index (ODI), radiculopathy, and Visual Analog Scale (VAS) for back pain and leg pain. Additionally, postoperative amelioration of central canal stenosis grade postoperatively was correlated with improved VAS scores [13,14].

Table 3

Correlation of postoperative MRI findings with clinical outcomes

Amelioration in the grade of central canal stenosis

Guen et al. developed an MRI-based grading system for central lumbar spinal stenosis [15]. Grade 0 is when there is an absence of LSS, Grade 1 represents mild stenosis, where each cauda equina remains distinctly separated; Grade 2 indicates moderate stenosis, with partial aggregation of the cauda equina; and Grade 3 denotes severe stenosis, where all cauda equina roots appear as a single bundle (Fig. 3) [15]. Only one study examined the association between the amelioration of the grade of central canal stenosis and clinical outcomes (Table 3) [16]. Amelioration of central canal stenosis grade postoperatively was correlated with improved VAS scores.

Fig. 3

Evaluation of the grading of central lumbar spinal stenosis.

Reversion of nerve root sedimentation sign

The nerve root sedimentation sign (SedSign), observed on axial MRI, was first described by Barz et al. [17]. In patients without LSS, MRI performed in the supine position typically shows sedimentation of the lumbar nerve roots toward the dorsal part of the dural sac due to gravity, a phenomenon defined as a negative SedSign. In LSS patients, the sedimentation of the nerve roots is absent, a finding designated as a positive SedSign (Fig. 4). Reversion of SedSign is defined as the conversion of a positive SedSign into a negative one.

Fig. 4

(A, B) Evaluation of the nerve roots sedimentation sign.

Three studies evaluated the relationship between postoperative reversion of SedSign and clinical outcomes (Table 3) [9,18,19]. Postoperative reversion of SedSign was significantly correlated with improvements in clinical outcomes in terms of VAS, ODI, and Zurich Claudication Questionnaire scores. Persistent postoperative positive SedSign is linked to poorer clinical outcomes. Therefore, postoperative negative SedSign with resolution of symptoms can be considered to be indicative of adequate decompression.

Negative redundant nerve roots

Redundant nerve roots (RNRs) were first described by Cressman and Pawl [20]. A positive finding of RNRs is defined as spinal stenosis due to increased epidural pressure. This increased pressure causes the cauda equina to become tangled, winding, coiled, and tortuous, resulting in a serpiginous appearance on sagittal T2-weighted image MRI (Fig. 5).

Fig. 5

Redundant nerve roots evaluation on sagittal T2-weighted image magnetic resonance imaging.

Two studies had examined the relationship between postoperative negative RNRs and clinical outcomes (Table 3) [12,21]. The conversion from positive to negative RNRs postoperatively is significantly correlated with improvement of clinical outcomes, whereas the persistence of postoperative RNRs is associated with poorer clinical outcomes.

Discussion

This systematic review of the best available evidence demonstrates that the postoperative MRI findings—particularly DSCA expansion, amelioration in the grade of central canal stenosis, reversion of SedSign, and negative RNRs—correlate with the clinical outcomes and reflect the adequacy and success of decompression surgery.

Adequate decompression is the most crucial factor in the surgical treatment of degenerative LSS. Postoperative measurement of DSCA expansion primarily reflects the degree of decompression achieved in the central canal, while providing limited assessment of the lateral recesses and foraminal stenosis. Only a few studies including those by Lee et al. [22], Siddiqui et al. [23], and Kim and Choi [24] have reported on DSCA expansion as observed on postoperative MRI. Their study revealed following decompression surgery, the cross-section area of the DSCA had been enlarged by 23%. Radiological diagnosis of spinal stenosis is typically established when the affected area is less than 100 mm², with an absolute value of 75 mm² being used as a diagnostic criterion for spinal stenosis [25]. Based on classification criteria, a DSCA greater than 100 mm² is considered normal, 76 to 100 mm² indicates moderate stenosis, while a DSCA less than 76 mm² is classified as severe stenosis [26]. This study indicates that a threshold of DSCA of <70 mm² after decompression surgery elevates the risk of persistent radiculopathy, while a DSCA of 70 mm² or more accompanied by a resolution of symptoms can be considered as indicative of adequate DSCA expansion, and thus adequate decompression. Study by Marie-Hardy et al. [27] suggested that DSCA expansion serves as an indicator of adequate decompression, indicating that significant DSCA enlargement reflects a broad and circumferential decompression.

LSS can be classified into central canal stenosis, lateral recess stenosis, and foraminal stenosis. This may explain why some studies found a correlation between expansion of postoperative DSCA and clinical improvement, while the others reported no such correlation [5]. Hermansen et al. [28] reported a postoperative DSCA expansion after surgical decompression, with an increase to more than twice the preoperative value. However, they did not associate the postoperative DSCA expansion with clinical improvement. Postoperative expansion of the DSCA following decompression surgery is due to the restoration of the original form of the dural sac under the action of intracapsular cerebrospinal fluid (CSF) pressure, relieving the pressure on the epidural vessels [11]. There was no correlation between postoperative DSCA and clinical improvement as assessed by functional outcomes in the study by Hermansen et al. [28]. They concluded that even a modest postoperative increase in DSCA can lead to significant symptom improvement, suggesting that less aggressive decompression may be sufficient. They observed that the mean baseline DSCA for the entire cohort was 52.0 mm² (standard deviation [SD]=21.2). This value increased to a mean of 117.2 mm² (SD=43.3) at the 3-month follow-up and further rose to 127.7 mm² (SD=51.4) after 2 years.

Lumbar spinal stenosis, particularly the central type, is defined when the CSF space on the anterior side is obliterated. Lee et al. [15] developed an MRI-based LSS grading system. According to Lee et al. [15], the stenosis grade is classified into four grades. Grade 0 is when there is an absence of LSS, Grade 1 represents mild stenosis, where each cauda equina remains distinctly separated; Grade 2 indicates moderate stenosis, with partial aggregation of the cauda equina; and Grade 3 denotes severe stenosis, where all cauda equina roots appear as a single bundle [15].

For lateral recess stenosis, the classification criteria are outlined below: (1) Grade 0: no contact between intervertebral disc and nerve root; (2) Grade 1: contact between intervertebral disc and nerve root with no deviation; (3) Grade 2: deviation of the nerve root caused by disc contact; (4) Grade 3: nerve root compression, characterized by a deformed nerve root.

For foraminal stenosis, the classification criteria are as shown below: (1) Grade 0: normal foramina with intact dorsolateral margin side of intervertebral disc; the foraminal epidural fat appears oval or inverted pear-shaped; (2) Grade 1: mild foraminal stenosis with deformation of epidural fat, though the residual fat still completely surrounds the exiting nerve root; (3) Grade 2: severe foraminal stenosis and deformation of the epidural fat, with the residual fat only partially surrounding the exiting nerve root; (4) Grade 3: advanced stenosis with complete obliteration of the epidural fat [26].

According to this review, postoperative improvement in the grade of central canal stenosis was associated with improved clinical outcomes. Therefore, grading of stenosis following decompression surgery can serve as a useful indicator for assessing the adequacy of the decompression.

Positive SedSign occurs when the nerve roots are displaced anteriorly within the thecal sac [29,30]. In degenerative LSS, particularly the central type, anterior displacement of the nerve root is typically caused by the hypertrophy of ligamentum flavum and arthropathy of the facet joints [18]. A positive SedSign may occur due to elevated epidural pressure at the level of LSS. This elevated pressure can lead to tethering of the nerve roots within the spinal canal, producing an appearance suggestive of mechanical entrapment or clamping [17]. A persistent positive SedSign after decompression surgery could indicate an inadequate surgical decompression or may suggest a surgical complication, such as the presence of intradural fat or an intradural cyst due to intraoperative dural tear. The appearance of a new positive SedSign following adequate decompression surgery may indicate a newly developed stenosis in a previously decompressed patient [9].

RNRs result from chronic mechanical compression within the spinal canal in LSS. The nerve roots are progressively displaced from the spinal canal due to the repeated movements of the lumbar spine under these circumstances. Over time, the nerve roots become coiled, elongated, and thickened, ultimately developing a serpentine appearance [21]. The conversion of preoperative positive RNRs to postoperative negative RNRs is significantly correlated with improvement of clinical outcomes, whereas the persistence of postoperative RNRs is associated with poorer clinical outcomes. This indicates that the presence of postoperative negative RNRs is indicative of adequate decompression.

It is important to review both normal and abnormal postoperative MRI findings, especially in patients with persistent postoperative symptoms or signs. The strength of this study is that it is the first systematic review to evaluate MRI findings as a means of assessing the adequacy of surgical decompression in LSS, with a correlation drawn between imaging results and clinical outcomes. However, although this review does not introduce novel MRI findings, it effectively consolidates and synthesizes the existing data on the subject. A limitation of this study is the lack of specification regarding the type of LSS, whether central, lateral recess, or foraminal stenosis. Furthermore, we were unable to perform the analysis on the strength of correlation between each specific postoperative MRI finding and clinical improvement. This limitation should be addressed in future studies to provide a more comprehensive understanding of the relationship between specific MRI findings and clinical outcomes.

Conclusions

Lumbar spinal stenosis is a clinical-radiological condition in which both clinical symptoms and MRI findings are crucial for initial assessment and evaluating postoperative outcomes. To evaluate the sufficiency of decompression surgery in LSS, the presence of postoperative MRI findings such as expansion of the DSCA, amelioration of the grade of stenosis, reversion of nerve roots SedSign, and negative RNRs are correlated with clinical improvement and should be carefully evaluated, especially by radiologists.

Article information Continued