Supraspinal Correlates of Patients with Chronic Pain Under Burst Spinal Cord Stimulation: A Proof-of-Concept Functional MRI Study Authors: Steffen Brenner, Achim Schilling, Patrick Krauss, Sherif Mohamed, Julia Fauth, Zohair Niroomand, Thomas Mehari Kinfe Published: 2026 ABSTRACT AbstractIntroductionChronic pain is associated with central pain network-associated alterations. Functional MRI (fMRI) provides a powerful, non-invasive imaging technique to visualize such alterations. This feasibility study attempted to investigate the impact of BurstDR spinal cord stimulation (SCS) on fMRI correlates in patients with chronic pain.MethodsThis proof-of-concept study included 11 patients (8 m/3 f; mean age 69.1 ± 11.2 years) with chronic back and/or leg pain with BurstDR SCS. Clinical outcomes were assessed at baseline and after 3 months BurstDR SCS (Numeric Rating Scale (NRS), Pittsburgh Sleep Quality Index (PSQI), Beck Depression Inventory (BDI), Oswestry Disability Index (ODI)). fMRI (1.5 T) was obtained using a classic block-design motor with task and rest with single-subject and group analyses.ResultsPain decreased at follow-up (NRS: 7.3 ± 1.7 vs. 4.4 ± 1.5; p < 0.001) with improved functional capacity (PSQI: 10.2 ± 3.6 vs. 8.1 ± 2.7; p = 0.009; BDI: 16.9 ± 8.2 vs. 11.9 ± 6.3; p = 0.027; ODI: 56.5 ± 18.3 vs. 45.2 ± 16.4; p = 0.001). In the fMRI analyses, ten patients were included. Movement of the left foot led to significant activation in the medial region of the right precentral gyrus (Z > 3.1, p < 0.05). No differences at the predefined threshold for the entire brain were detected, although the uncorrected voxel-wise analysis showed several regions with increased activation post BurstDR SCS (bilateral middle temporal gyri (left: Z = 3.56, uncorrected p < 0.001; right: Z = 4.48, uncorrected p < 0.0001)). Functional connectivity analyses in 9 patients revealed no significant differences in task-modulated functional connectivity.ConclusionsBurstDR SCS reduced pain intensity and improved functional status associated with task-related activation of the motor cortex, which was reproducible across the entire cohort. Further randomized controlled studies are warranted to validate these findings and to elucidate the underlying mechanisms.Trial RegistrationGerman Clinical Trials Register (DRKS ID: DRKS00035826).Keywords: Chronic pain, Burst spinal cord stimulation, Functional MRI, Objective outcome measures Key Summary Points Why carry out this study? Spinal cord stimulation (SCS) for chronic pain lacks objective outcome measures.Predominantly applied subjective score-based outcome measures are biased by intra- and inter-individual variabilities.To determine the effects of BurstDR SCS on structural and functional central circuits using functional MRI (fMRI). What was learned from the study? Our preliminary findings indicate that BurstDR SCS promotes pain relief, while changes in distinct brain regions quantified by task-related fMRI require further investigations.Large-scale, sham-controlled trials are needed to clearly characterize these clinical effects on both pain and fMRI. Introduction Spinal cord stimulation (SCS) is an established and widely used therapy for patients with chronic pain refractory to conservative treatment. First, low-frequency (tonic) SCS was introduced and several studies have shown effectiveness [1–3]. However, the development of paresthesia-free sub-perception stimulation paradigms improved treatment options by enhancing therapeutic response rates. These paradigms include BurstDR and high-frequency stimulation. Clinical efficacy and superiority to tonic SCS have been demonstrated in several randomized controlled trials, although a considerable proportion of patients with pain remain responsive to tonic SCS [1, 4, 5]. Despite these robust clinical data, the precise mechanisms underlying SCS remain incompletely understood. This lack of mechanistic insight poses a significant limitation, as it hampers optimized patient selection, individualized therapy adjustment, and objective treatment monitoring. Consequently, substantial efforts have been made to elucidate the biological and neurophysiological mechanisms of SCS. Electrophysiological studies have provided substantial insight into the mechanisms of SCS. Low-frequency suprathreshold tonic stimulation has been shown to activate Aβ-fiber collaterals within the dorsal columns, resulting in central neurotransmitter release—predominantly gamma-aminobutyric acid (GABA)—in the dorsal horn via segmental and supraspinal pathways. In addition, antidromic activation in peripheral nerves contributes to segmental peripheral effects [6]. However, the mechanisms of BurstDR and high-frequency stimulation appear to differ and are less well characterized. High-frequency SCS is thought to reduce pain by modulating hyperactive wide dynamic range (WDR) neurons and central sensitization [7]. It was thought that high-frequency stimulation may produce a reversible conduction block in the axons of the dorsal column; however, recent electrophysiological studies challenge this interpretation. It is more likely that pain relief is achieved through desynchronization of dorsal column neuronal networks [8]. BurstDR stimulation primarily modulates activity within the spinal dorsal horn and appears to act via non-GABAergic mechanisms [7]. Additionally, increasing evidence indicates that the analgesic effects of BurstDR SCS involve modulation of supraspinal pain processing networks, including the pregenual anterior cingulate cortex, posterior cingulate cortex, and parahippocampus [9, 10]. Beyond spinal electrophysiology, molecular biological approaches have been employed to identify additional pathways of action. Particularly for BurstDR changes of cytokines have been observed [11, 12]. It has been demonstrated that BurstDR increases levels of the anti-inflammatory cytokine interleukin-10, which is thought to contribute not only to analgesia but in addition to the promotion of nerve fiber regeneration [7]. This suggests another possible mechanism of action for BurstDR. Chronic pain is known to induce both functional and structural alterations within the brain [13–15]. Functional magnetic resonance imaging (fMRI) provides a powerful, non-invasive technique for visualizing such brain changes and may serve as a potential biomarker for chronic back and/or leg pain [16, 17]. Increasing evidence indicates that SCS induces measurable changes at the cortical and subcortical level, with paradigm-dependent effects [6]. Tonic and BurstDR SCS both modulate the lateral ascending pain pathway and the descending inhibitory pain transmission, but BurstDR SCS has been suggested to additionally influence the affective medial pain pathway, particularly involving the dorsal anterior cingulate cortex [10]. Several studies confirmed these findings linking BurstDR effects to the sensorimotor areas and to the emotional–affective components of pain [18–20]. This study primarily aimed to visualize the effects of BurstDR SCS on sensorimotor brain regions. As a proof-of-concept study, its objective was to explore whether SCS therapy with BurstDR stimulation is associated with task-related fMRI changes during motor paradigms in patients with chronic back and/or leg pain, along with analysis of secondary endpoints comprising of subjective clinical outcome measures. This approach was chosen because it represents a simple and non-invasive measurement method in fMRI, and could therefore also be applied in clinical settings in the future. Methods This single-center study included patients with chronic, refractory back and/or leg pain. The study protocol was approved by an independent internal local ethics committee (IRB No. 2024–671), conducted in accordance with the principles of the latest revision of the Declaration of Helsinki, and registered on January 21, 2025, in the German Clinical Trials Register (DRKS ID: DRKS00035826). Written informed consent was obtained from all participants. Participants Eleven patients eligible for SCS were recruited prospectively between January and October 2025. All participants were reviewed by an interdisciplinary pain board and defined as refractory to pharmacological and/or behavioral and/or alternative surgical therapies suffering from chronic leg and/or back pain. Refractory to pharmacological treatment was defined as the use of at least one opioid and one additional medication for neuropathic pain. This medication regimen was maintained at the time of SCS implantation. Additionally, participants were required to maintain stable pain medication for at least four weeks prior to SCS implantation and throughout the 3-month study period. Pain intensity was assessed using the Numeric Rating Scale (NRS). Sleep quality was evaluated with the Pittsburgh Sleep Quality Index (PSQI), and depressive symptoms were assessed using the Beck Depression Inventory (BDI). The level of disability was determined using the Oswestry Disability Index (ODI). All outcome measures used in this study (NRS, BDI, ODI, and PSQI) are freely available for research purposes. The scores were obtained prior to SCS implantation and at the 3-month follow-up. At the same time, all participants underwent fMRI. One participant was excluded from further analysis due to the amputation of the left lower extremity between sessions. This prevented the motor task from being repeated during the subsequent follow-up scan. Consequently, data from ten participants were incorporated into the final analyses (Fig. 1). CONSORT-style flow chart for the trial with inclusion/exclusion criteria. Initially, 11 patients were measured. One patient had to be excluded for the follow-up measurement resulting in ten patients analyzed in the motor task analyses. For the functional connectivity analyses, another patient had to be excluded due to shorter task blocks Protocol for Implantation Technique and Trial Period SCS electrode implantation targeted the thoracic spinal cord levels between Th7 and Th11 in each participant. A single 16-contact surgical paddle-type lead (Penta; Abbott, Plano, TX, USA) was implanted under general anesthesia. The paddle lead was connected to an extension cable and externalized in all participants to allow a 7-day trial stimulation period. Stimulation was delivered using BurstDR (patented by Abbott) with a frequency of 40 Hz, a pulse width of 1000 µs, and an intra-burst frequency of 500 Hz. Patients who achieved at least a 30% reduction in pain intensity compared with baseline during the trial phase subsequently underwent implantation of a permanent pulse generator (Eterna; Abbott, Plano, TX, USA) in a second procedure. Electrode configurations and stimulation amplitudes were individually optimized to maintain effective pain relief while operating at sub-perception thresholds (the mean sub-perception threshold applied was 60%). During MRI scans, the system was switched to MRI mode, and stimulation was temporarily deactivated to eliminate electromagnetic interference and to comply with the manufacturer's safety guidelines. MRI Data Acquisition All imaging was performed on a 1.5-Tesla MAGNETOM Aera scanner (Siemens Healthcare, Erlangen, Germany) with a standard 12-channel head coil. Functional images were acquired using a T2-weighted gradient-echo echo-planar imaging (EPI) sequence with the following parameters: TR = 4000 ms, TE = 50 ms, flip angle = 90°, voxel size = 3.9 × 3.9 × 4.0 mm3, 60 volumes (total duration = 240 s), interleaved slice acquisition, and acceleration factor of 2. A 3D structural T1-weighted MPRAGE sequence was obtained for anatomical reference and for registration (TR = 2000 ms, TE = 3.72 ms, voxel size = 1.0 × 1.0 × 1.0 mm3). In addition, a dual-echo gradient echo field map was acquired to correct for magnetic field inhomogeneities (TR = 517 ms, TE1 = 4.76 ms, TE2 = 9.52 ms). A classic block-design motor with a task was used. Participants were asked to perform a repeated arbitrary movement of the left foot during 40-s “movement blocks” alternating with 40-s “rest blocks,” beginning with rest, for a total duration of 240 s (60 volumes per session). For one patient, one of the sessions consisted of shorter 24-s blocks; this was taken into account in the analysis. Participants were visually monitored to ensure compliance with the task instructions. The standard DICOM data exported from the scanner was converted to NIfTI format using dcm2niix [21]. After brain extraction using HD-BET [22], data were pre-processed with FEAT (FMRI Expert Analysis Tool, [23]) version 6.00, part of FSL (FMRIB's Software Library, http://www.fmrib.ox.ac.uk/fsl). The following preprocessing steps were implemented: Head motion was corrected using MCFLIRT with six degrees of freedom. Motion parameters (three translations and three rotations) were estimated for each volume and included as nuisance regressors in the statistical model. Visual inspection of motion plots and realigned images was performed to ensure data quality. No datasets were excluded based on head motion. Spatial smoothing was applied using a Gaussian kernel with a full width at half maximum (FWHM) of 5 mm. High-pass temporal filtering was applied using a cutoff period of 90 s to remove low-frequency drifts. Functional series were registered to structural images using FLIRT with boundary-based registration (BBR), and nonlinear normalization to MNI152 space was performed using FNIRT (Figs. 1, 2). Overview demonstrating the protocol of MRI data acquisition and analysis. A classical block-design motor task was applied (40-s movement alternating with 40-s rest). Data were converted, brain-extracted, and preprocessed, including motion correction, fieldmap-based unwarping, spatial smoothing, high-pass filtering, registration to structural images, and nonlinear normalization Functional Connectivity Analysis One patient was excluded from the connectivity analyses because one of their sessions used shorter task blocks (24 s active/24 s rest), which provided insufficient data for the reliable estimation of functional connectivity. The connectivity analyses therefore included nine patients (18 sessions) and were performed using CONN Toolbox [24] release 22.v2407 and SPM release 25.25.01.02 [25] (Fig. 1). Task-modulated functional connectivity was examined using generalized psychophysiological interaction model [26, 27]. Six regions of interest (ROIs) were selected from the Harvard–Oxford atlas [28], including bilateral precentral gyrus, supplementary motor area, and bilateral thalamus. For each pair of ROIs, gPPI models included the seed BOLD time series (physiological factor), task condition regressors convolved with the canonical hemodynamic response function (psychological factors), and their interaction terms. ROI-to-ROI connectivity estimates were derived from the regression coefficients of the interaction terms. Group-level analyses tested for within-subject pre- to post-implantation changes using a paired samples general linear model with subject-specific regressors to account for repeated measures. Inferences were performed at the level of individual clusters (groups of contiguous voxels). Cluster-level inferences were based on parametric statistics from Gaussian random field theory. Results were thresholded using a combination of a cluster-forming p < 0.001 voxel-level threshold, and a familywise corrected p-FDR < 0.05 cluster-size threshold. Statistical Analysis Patient demographics are presented as mean values ± standard deviation (SD) or as relative frequencies and percentages. The outcomes are reported using mean values ± SD. The normality of data distribution was assessed using the Kolmogorov–Smirnov test. As normality was met, a paired t test was performed. Single-subject analyses of the fMRI were performed using a general linear model (GLM) as implemented in FSL FEAT. Task conditions (movement and rest) were modeled using boxcar regressors convolved with a gamma hemodynamic response function. Temporal derivatives were included to account for temporal variability in the hemodynamic response. Parameter estimates were obtained using FILM (FMRIB's Improved Linear Model) with local autocorrelation correction. The primary contrast of interest was task > rest. Statistical inference was performed at the whole-brain level using cluster-based correction for multiple comparisons based on Gaussian random field theory. A cluster-defining threshold of Z > 3.1 was applied at the voxel level, and clusters were considered statistically significant at a family-wise error–corrected threshold of p < 0.05. Results BurstDR SCS Effects on Pain Levels and Functional Impairment Eleven patients (8 men [73%], 3 women [27%]; mean age 69.1 ± 11.2 years) met the inclusion criteria and demonstrated a successful SCS trial period, leading to permanent spinal cord stimulator implantation. Table 1 summarizes baseline characteristics and applied SCS stimulation patterns. Table 1Distribution of patient characteristics at baseline, including demographics, pain intensity, functional state, spine level of SCS implantation, stimulation parameters, and tolerance threshold valuesNumberGender (age)Spine levelActive contactsFrequency (Hz)Intra-burst-frequency (Hz)Pulse width (µs)Amplitude (mA)DiagnosisPre NRSPost NRSPre BDIPost BDIPre ODIPost ODIPre PSQIPost PSQI1M (71)Th7/811-, 12-, 9+4050010000.4PSPS7518156256762M (52)Th7/84-, 6+4050010000.7PSPS942715744816103M (75)Th81-, 2+4050010001.5PSPS77191975.675.611114M (75)Th96-, 2+4050010000.9PAD62151550281165M (60)Th7/87-, 10+4050010000.45PSPS6413124236776F (56)Th72-, 5-, 6-, 10+, 13+, 14+, 16+4050010000.3PSPS1062719625013117M (60)Th7/811-, 13+4050010000.45PSPS968064481068F (64)Th710-, 6+, 8+4050010000.7PSPS84299806014119F (81)Th1012-, 4+4050010000.8PAD7415154033.310810M (84)Th10/116-, 5+4050010000.55PAD433218163311M (82)Th73+, 5-4050010000.35PSPS73121054461010M male, F female, Th thoracic, Hz Hertz, µs microseconds, mA milliampere, PSPS persistent spinal pain, PAD peripheral arterial disease, NRS numeric rating scale, BDI Beck Depression Inventory, ODI Oswestry Disability Scale, PSQI Pittsburgh Sleep Quality Inventory Distribution of patient characteristics at baseline, including demographics, pain intensity, functional state, spine level of SCS implantation, stimulation parameters, and tolerance threshold values M male, F female, Th thoracic, Hz Hertz, µs microseconds, mA milliampere, PSPS persistent spinal pain, PAD peripheral arterial disease, NRS numeric rating scale, BDI Beck Depression Inventory, ODI Oswestry Disability Scale, PSQI Pittsburgh Sleep Quality Inventory Pain intensity decreased significantly at the 3-month follow-up compared with baseline (mean NRS baseline: 7.3 ± 1.7 vs. 3-month follow-up: 4.4 ± 1.5; p < 0.001). Sleep quality improved significantly following SCS (mean PSQI baseline: 10.2 ± 3.6 vs. 3-month follow-up: 8.1 ± 2.7; p = 0.009). Mood disturbance was significantly reduced (mean BDI baseline: 16.9 ± 8.2 vs. 3-month follow-up: 11.9 ± 6.3; p = 0.027). Disability related to pain also improved significantly at 3 months (mean ODI baseline: 56.5 ± 18.3 vs. 3-month follow-up: 45.2 ± 16.4; p = 0.001) (Fig. 3). Subgroup analyses of patients with persistent spinal pain syndrome (PSPS) revealed similar results: Pain intensity decreased significantly (mean NRS baseline: 7.9 ± 1.4 vs. 3-month follow-up: 4.9 ± 1.4; p = 0.001). Sleep quality improved significantly following SCS (mean PSQI baseline: 11.0 ± 3.2 vs. 3-month follow-up: 9.0 ± 2.3; p = 0.037). Mood disturbance was significantly reduced (mean BDI baseline: 19.1 ± 7.9 vs. 3-month follow-up: 12.4 ± 6.2; p = 0.026). Disability related to pain also improved significantly at 3 months (mean ODI baseline: 64.2 ± 12.4 vs. 3-month follow-up: 52.5 ± 11.7; p = 0.006). For the patients with peripheral artery disease (PAD), the following results were obtained (means ± standard deviations are reported, as formal statistical testing is not meaningful given the small sample size of n = 3): Pain intensity decreased (mean NRS baseline: 5.7 ± 1.5 vs. 3-month follow-up: 3.0 ± 1.0). Sleep quality improved (mean PSQI baseline: 8.0 ± 4.4 vs. 3-month follow-up: 5.7 ± 2.5). Mood disturbance was slightly reduced (mean BDI baseline: 11.0 ± 6.9 vs. 3-month follow-up: 10.7 ± 7.5). Disability related to pain also improved (mean ODI baseline: 36.0 ± 16.4 vs. 3-month follow-up: 25.8 ± 8.9). Numeric rating scale for pain (NRS), sleep quality (PSQI = Pittsburgh Sleep Quality Index), mood assessment (BDI = Beck Depression Inventory), and disability related to pain (ODI = Oswestry Disability Index): a comparison of baseline assessment and after 3 months BurstDR spinal cord stimulation (SCS). The box plots show median values with interquartile range (Q1–Q3), 1.5 interquartile range (whiskers), outliers and p values. There was a significant improvement in all parameters after 3 months BurstDR SCS. * Indicates p values < 0.05 (statistically significant) Task-Related Activation Before and After SCS Implantation In the pre-implantation group analysis, movement of the left foot led to significant activation in the medial region of the right precentral gyrus (Z > 3.1, p < 0.05), which is consistent with the expected representation of the lower extremities within the primary motor cortex (Fig. 4a). Similarly, in the post-implantation group analysis, the motor task again resulted in significant activation in the right medial precentral gyrus (Z > 3.1, p < 0.05). The activation clusters corresponded spatially with the pre-implantation results (Fig. 4b). These results remained stable in the subgroup analyses of patients with PSPS (Z > 3.1, p < 0.05). There was no subgroup analysis for patients with PAD, because of very small sample size (n = 2). Group mean activation during left-foot movement; a before spinal cord stimulation (SCS) implantation; b after SCS implantation. Significant activation was observed in the medial region of the right precentral gyrus, respectively (Z > 3.1, p < 0.05). The activation clusters showed spatial correspondence between the pre- and post-implantation results. The sagittal, coronal, and axial planes are shown, respectively Paired Comparison (Before vs. After Implantation) and Functional Connectivity Analysis The paired t test comparing activation before and after implantation revealed no significant differences at the predefined threshold for the entire brain (i.e., no clusters were identified). This did not change even after lowering the CDT to Z > 2.3. In the uncorrected voxel-wise analysis several regions showed increased activation in the post-implantation condition, most notably in the bilateral middle temporal gyri (left: Z = 3.56, uncorrected p < 0.001; right: Z = 4.48, uncorrected p < 0.0001). However, none of these effects remained significant after applying cluster-based correction for multiple comparisons. These results remained stable in the subgroup analysis of patients with PSPS. There was no subgroup analysis for patients with PAD, because of very small sample size (n = 2). ROI-to-ROI connectivity analyses revealed no significant differences in task-modulated functional connectivity between pre- and post-implantation scans after correction for multiple comparisons (all FDR-corrected p > 0.05). Exploratory inspection of uncorrected results showed inconsistent connectivity changes across subjects and ROI pairs; however, these effects did not remain statistically significant after FDR correction. In the subgroup analysis of patients with PSPS ROI-to-ROI connectivity analyses revealed no significant differences in task-modulated functional connectivity between pre- and post-implantation scans after correction for multiple comparisons (all FDR-corrected p > 0.05). There was no subgroup analysis for patients with PAD, because of very small sample size (n = 2). Discussion In this study, patients with back and/or leg pain, who were refractory to conservative treatment, were treated with BurstDR SCS. We proved a good responsiveness to BurstDR in terms of pain reduction and improved functional outcome, which is consistent with previous studies related to BurstDR SCS [9, 30]. Pre-existing disturbed sleep quality and clinical depressive symptoms as well as functional deficits due to pain were present in our study cohort. We attempted to visualize effects of BurstDR SCS in the brain using fMRI. We chose to target changes in sensorimotor brain regions, as modulation of these areas has been proposed as one potential mechanism of action [18–20]. Task-related activation of the motor cortex was robust and reproducible across the entire cohort, and no systematic changes in cortical activation were observed 3 months after SCS implantation. Functional connectivity analyses revealed no significant differences in task-modulated functional connectivity after treatment with BurstDR SCS. Thus, in this feasibility study, we were not able to demonstrate robust fMRI changes directly linked to pain modulation at the group level. Previously published data is only available for tonic SCS, except for one study [31] in which four frequencies (4 Hz, 60 Hz, 500 Hz, and 1 kHz) with both subthreshold and suprathreshold amplitudes were compared. All of these studies examined only a limited number of patients. Studies revealed complex and heterogeneous patterns of brain activation and deactivation during SCS, with substantial variation in regions affected and directionality of changes. Some reports found changes in sensorimotor [31–33] and thalamic regions [31, 34, 35]. In this study, sensorimotor changes could not be detected. Additionally, multiple studies demonstrated involvement of limbic and emotional processing circuits [31, 32, 34–36]. Four studies reported on changes in the frontal and association cortex [31–33, 35]. Furthermore, De Groote et al. [31] documented activation of the basal ganglia and claustrum bilaterally during suprathreshold SCS. Moens et al. [34] found a decrease of activity with pain relief due to SCS in the inferior olivary nucleus and cerebellum. Deogaonkar et al. [33] observed decreased activation of default mode network (DMN), such as the precuneus. Several studies examined relationships between brain activity changes and clinical pain outcomes. As already mentioned, Moens et al. [34] demonstrated that immediate pain relief obtained by SCS correlated negatively with activity in the inferior olivary nucleus, cerebellum, and rostral anterior cingulate cortex. The bilateral medial thalamus emerged as a key mediator, with deactivation proposed as a marker of therapeutic response. Rasche et al. [35] found that activation of specific brain areas (cingulate gyrus, thalamus, prefrontal cortex, supplementary motor area, and postcentral gyrus) occurred during high pain levels and reduced after pain reduction, indicating an immediate relationship between SCS effects and brain activity changes. Deogaonkar et al. [33] reported decreased connectivity between several regions related to pain perception, including the left frontal insula, right primary and secondary somatosensory cortices, as well as in regions involved in the DMN, such as the precuneus. Pahapill et al. [36] observed that patients with longer therapeutic effects of SCS had fewer alterations in their connectivity patterns, suggesting that sustained clinical benefit may be associated with more normalized brain network organization. The mentioned reports used heterogenous study designs. The studies employed task and resting-state paradigms with and without block designs. Most of the examinations were obtained with externalized SCS systems and with “on” and “off” conditions. However, none of the studies provides preoperative data. The patients’ pain syndromes were also heterogeneous and included PSPS, complex regional pain syndrome and other neuropathic pain syndromes. Given this heterogeneity, meaningful comparisons with the present study are hardly feasible. None of the mentioned investigated BurstDR stimulation. Limitations Obviously, this pilot study has several limitations that make it difficult to draw conclusions as in the mentioned studies. First, the relatively small number of cases may have limited the statistical power to detect changes in functional activation; consequently, the null fMRI finding is not readily interpretable, as it may reflect either a true absence of effect or insufficient statistical power. In addition, the 3-month interval between SCS implantation and follow-up may have been insufficient to detect potential long-term neuroplastic changes. This is of particular relevance, as during MRI acquisition, the system was set to MRI mode and stimulation was temporarily deactivated to prevent electromagnetic interference and to comply with the manufacturer’s safety guidelines. Consequently, fMRI data were obtained without active stimulation at both time points. The study design therefore does not allow assessment of acute neural effects of BurstDR, but instead captures potential carry-over or neuroplastic changes following 3 months of stimulation. Future studies with larger cohorts and longer follow-up periods may be able to detect functional adaptations of the brain following BurstDR SCS treatment in patients with chronic back and/or leg pain. Furthermore, the functional connectivity analysis was only feasible to a limited extent due to the study design, including the use of block-based analyses and the applied imaging parameters (1.5 Tesla field strength, long repetition time, and only 60 measurements per session). Future studies should therefore employ resting-state fMRI at 3 Tesla with a higher number of acquisitions. All patients included in this study suffered from back and/or leg pain, but the underlying diagnosis differed. Pain was predominantly due to PSPS but also to PAD in three patients. This could have additionally influenced the analysis, as different diagnoses may be associated with distinct functional changes in the brain. This study aimed to visualize the effects of BurstDR SCS on the sensorimotor areas of the brain, as modulation of these areas has been proposed as one potential mechanism of action [18–20]. However, it is conceivable that activation of the affective medial pain pathway, particularly involving the dorsal anterior cingulate cortex, may be more prominent and might therefore have yielded a stronger signal [10]. Lastly, this study did not include healthy controls, nor did it include additional SCS waveforms, such as tonic and/or cyclic stimulation. Thus, the study cannot distinctly distinguish between treatment-related effects and potential time-dependent changes, learning effects, or scanner variability. For the BOLD signal relative changes are detected. It is not possible to determine whether the changes measured in this study differ from those in healthy controls, as there are no reference values. This is because different MR scanners, field strengths, head coils, etc. influence the quality and sensitivity of the BOLD signal. Comparisons are therefore only meaningful if healthy controls are examined on the same MR scanner using the same paradigms. Conclusions Currently, no objective and reliable biomarkers exist to predict SCS outcomes for all stimulation waveforms and neuroanatomical targets. Previous research has predominantly relied on biological and neurophysiological methods to address this issue. fMRI provides a powerful, non-invasive imaging technique to visualize functional changes in the brain and thus it may be capable of visualizing changes in the brain induced by BurstDR SCS. This feasibility study attempted to investigate the impact of BurstDR SCS on fMRI changes in patients with chronic back and/or leg pain. But no changes have been detected. Data on functional brain changes are available, particularly for tonic SCS. However, these findings are not yet conclusive. To our knowledge this is the first study in which fMRI examinations of patients undergoing SCS with BurstDR stimulation were performed and preoperative baseline data were compared with postoperative data. Task-related activation of the motor cortex was robust and reproducible across the entire cohort, but functional connectivity analyses failed to link stimulation effects to fMRI changes. Further studies with healthy and/or sham-control groups are needed to elucidate fMRI changes due to SCS. Acknowledgements We thank the participants of the study. Medical Writing/Editorial Assistance No medical writing or editorial assistance was received during the preparation of this article. Author Contributions All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Thomas Mehari Kinfe, Steffen Brenner, Julia Fauth, Sherif Mohamed, and Zohair Niroomand. The first draft of the manuscript was written by Thomas Mehari Kinfe, Steffen Brenner, and Zohair Niroomand. Achim Schilling, Patrick Krauss, and Sherif Mohamed reviewed and edited the manuscript. All authors read and approved the final manuscript. Funding No funding or sponsorship was received for this study or publication of this article. Data Availability The datasets generated during and/or analyzed during the current study are available from the corresponding authors on reasonable request. Declarations Conflict of Interest Thomas Mehari Kinfe, MD, PhD, works as a consultant for Abbott Inc. and has been paid for presentation and received conference travel support from Abbott Inc. Thomas Mehari Kinfe is the founder and CEO of NAGUS Consulting and is an Editorial Board member of Pain and Therapy. Thomas Mehari Kinfe was not involved in the selection of peer reviewers for the manuscript nor any of the subsequent editorial decisions. Steffen Brenner has nothing to disclose. Achim Schilling has nothing to disclose. Patrick Krauss has nothing to disclose. Sherif Mohamed has nothing to disclose. Julia Fauth has nothing to disclose. Zohair Niroomand has nothing to disclose. Ethical Approval The study protocol was approved by an independent internal local ethics committee (IRB No. 2024-671), conducted in accordance with the principles of the latest revision of the Declaration of Helsinki, and registered on January 21, 2025, in the German Clinical Trials Register (DRKS ID: DRKS00035826). Written informed consent was obtained from all participants. Footnotes Zohair Niroomand and Thomas Mehari Kinfe contributed equally to this work. References Associated Data Data Availability Statement The datasets generated during and/or analyzed during the current study are available from the corresponding authors on reasonable request.