Predicting the Response of High Frequency Spinal Cord Stimulation in Patients with Failed Back Surgery Syndrome: A Retrospective Study with Machine Learning Techniques

The Role and Applications of Artificial Intelligence in the Treatment of Chronic Pain

PURPOSE OF REVIEW

This review aims to explore the interface between artificial intelligence (AI) and chronic pain, seeking to identify areas of focus for enhancing current treatments and yielding novel therapies.

RECENT FINDINGS

In the United States, the prevalence of chronic pain is estimated to be upwards of 40%. Its impact extends to increased healthcare costs, reduced economic productivity, and strain on healthcare resources. Addressing this condition is particularly challenging due to its complexity and the significant variability in how patients respond to treatment. Current options often struggle to provide long-term relief, with their benefits rarely outweighing the risks, such as dependency or other side effects. Currently, AI has impacted four key areas of chronic pain treatment and research: (1) predicting outcomes based on clinical information; (2) extracting features from text, specifically clinical notes; (3) modeling 'omic data to identify meaningful patient subgroups with potential for personalized treatments and improved understanding of disease processes; and (4) disentangling complex neuronal signals responsible for pain, which current therapies attempt to modulate. As AI advances, leveraging state-of-the-art architectures will be essential for improving chronic pain treatment. Current efforts aim to extract meaningful representations from complex data, paving the way for personalized medicine. The identification of unique patient subgroups should reveal targets for tailored chronic pain treatments. Moreover, enhancing current treatment approaches is achievable by gaining a more profound understanding of patient physiology and responses. This can be realized by leveraging AI on the increasing volume of data linked to chronic pain.

Clinical outcomes of spinal cord stimulation in patients with intractable leg pain in Japan

BACKGROUND

Neuromodulation through spinal cord stimulation (SCS) is a therapeutic option for relieving leg pain and improving the chances of limb salvage in patients with intractable chronic limb-threatening ischemia (CLTI); however, there is no consensus on its indications.

OBJECTIVE

The aim of this study was to assess the clinical outcomes of SCS in patients with intractable leg pain caused by various diseases treated in the department of cardiovascular medicine in Japan.

METHODS

This was a retrospective study of patients who underwent SCS for pain management. Patients were considered eligible for the therapy if they met the following criteria: (1) intractable leg pain (numerical rating scale [NRS] score of 10), (2) no revascularization option, and (3) no septicemia.

RESULTS

Twenty patients (mean age: 77 years; men/women: 11/9) were included in this study. The NRS score of the patients significantly reduced from 10 ± 0 before procedure to 4 ± 3 at discharge (p < 0.001). The clinical response rate of the entire cohort was 65% (13/20) at 17 ± 14 months after implantation; however, patients with intractable CLTI showed a low response rate (45%), whereas those with subacute limb ischemia showed a high response rate (100%). A multivariable regression analysis showed that hemoglobin level was significantly associated with treatment response, even after adjusting for age and sex (p = 0.026). The area under the receiver operating characteristic curve for the correlation between hemoglobin level (cutoff, 11.4 g/dL) and clinical response to SCS was 0.824 (0.619-1).

CONCLUSIONS

SCS can reduce clinical symptoms in majority of patients with intractable leg pain. Although implantation of an SCS device has been shown to improve microvascular perfusion insufficiency, the correlation between hemoglobin level and the clinical effect of SCS indicates that a preserved microcirculatory vascular bed is essential for the therapy to be effective.

Variables associated with nonresponders to high-frequency (10 kHz) spinal cord stimulation

INTRODUCTION

The use of spinal cord stimulation (SCS) therapy to treat chronic pain continues to rise. Optimal patient selection remains one of the most important factors for SCS success. However, despite increased utilization and the existence of general indications, predicting which patients will benefit from neuromodulation remains one of the main challenges for this therapy. Therefore, this study aims to identify the variables that may correlate with nonresponders to high-frequency (10 kHz) SCS to distinguish the subset of patients less likely to benefit from this intervention.

MATERIALS AND METHODS

This was a retrospective single-center observational study of patients who underwent 10 kHz SCS implant. Patients were divided into nonresponders and responders groups. Demographic data and clinical outcomes were collected at baseline and statistical analysis was performed for all continuous and categorical variables between the two groups to calculate statistically significant differences.

RESULTS

The study population comprised of 237 patients, of which 67.51% were responders and 32.49% were nonresponders. There was a statistically significant difference of high levels of kinesiophobia, high self-perceived disability, greater pain intensity, and clinically relevant pain catastrophizing at baseline in the nonresponders compared to the responders. A few variables deemed potentially relevant, such as age, gender, history of spinal surgery, diabetes, alcohol use, tobacco use, psychiatric illness, and opioid utilization at baseline were not statistically significant.

CONCLUSION

Our study is the first in the neuromodulation literature to raise awareness to the association of high levels of kinesiophobia preoperatively in nonresponders to 10 kHz SCS therapy. We also found statistically significant differences with greater pain intensity, higher self-perceived disability, and clinically relevant pain catastrophizing at baseline in the nonresponders relative to responders. It may be appropriate to screen for these factors preoperatively to identify patients who are less likely to respond to SCS. If these modifiable risk factors are present, it might be prudent to consider a pre-rehabilitation program with pain neuroscience education to address these factors prior to SCS therapy, to enhance successful outcomes in neuromodulation.

Healthcare Utilization (HCU) Reduction with High-Frequency (10 kHz) Spinal Cord Stimulation (SCS) Therapy

Spinal cord stimulation (SCS) is a well-established treatment for patients with chronic pain. With increasing healthcare costs, it is important to determine the benefits of SCS in healthcare utilization (HCU). This retrospective, single-center observational study involved 160 subjects who underwent implantation of a high-frequency (10 kHz) SCS device. We focused on assessing trends in HCU by measuring opioid consumption in morphine milligram equivalents (MME), as well as monitoring emergency department (ED) and office visits for interventional pain procedures during the 12-month period preceding and following the SCS implant. Our results revealed a statistically significant reduction in HCU in all domains assessed. The mean MME was 51.05 and 26.52 pre- and post-implant, respectively. There was a 24.53 MME overall decrease and a mean of 78.2% statistically significant dose reduction (p < 0.0001). Of these, 91.5% reached a minimally clinically important difference (MCID) in opioid reduction. Similarly, we found a statistically significant (p < 0.01) decrease in ED visits, with a mean of 0.12 pre- and 0.03 post-implant, and a decrease in office visits for interventional pain procedures from a 1.39 pre- to 0.28 post-10 kHz SCS implant, representing a 1.11 statistically significant (p < 0.0001) mean reduction. Our study reports the largest cohort of real-world data published to date analyzing HCU trends with 10 kHz SCS for multiple pain etiologies. Furthermore, this is the first and only study evaluating HCU trends with 10 kHz SCS by assessing opioid use, ED visits, and outpatient visits for interventional pain procedures collectively. Preceding studies have individually investigated these outcomes, consistently yielding positive results comparable to our findings.

Correlation between spinal cord stimulation analgesia and cortical dynamics in pain management

AIM

Spinal cord stimulation (SCS) is an effective method to treat neuropathic pain. It is necessary to identify the responders of SCS analgesia before implantation. The aim of this study is to investigate the relationship between the cortical dynamics and SCS analgesia responders in pain management.

METHODS

Resting-state EEG recording was performed in patients who underwent short-term implantation of spinal cord stimulation for pain therapy. We then did spectral analysis to capture the pattern of cortical oscillation between neuromodulation therapy analgesia responders and nonresponders.

RESULTS

About 58.3% (14 out of 24) of participants were considered as analgesia responders, with average visual analogue scores reduction of 4.8 ± 1.0 after surgery, and 2.1 ± 0.7 for the nonresponder subgroup, respectively. The alpha oscillation was significantly enhanced in responder cohort compared with nonresponders. We also observed an increasing spectral power of gamma band in responders. Furthermore, the attenuation of pain severity was significantly correlated with the global alpha oscillation activity (r = 0.60, P = 0.002). Likely, positive and significant correlation was found between the pain relief and gamma activity (r = 0.58, P = 0.003).

CONCLUSIONS

Distinct pattern of neural oscillation is associated with the analgesic effect of spinal cord stimulation in pain management, enhancement of cortical alpha and gamma oscillation may be a predictor of analgesia responders.

Machine Learning in Spinal Cord Stimulation for Chronic Pain

Spinal cord stimulation (SCS) is an effective treatment for chronic neuropathic pain. The success of SCS is dependent on candidate selection, response to trialing, and programming optimization. Owing to the subjective nature of these variables, machine learning (ML) offers a powerful tool to augment these processes. Here we explore what work has been done using data analytics and applications of ML in SCS. In addition, we discuss aspects of SCS which have narrowly been influenced by ML and propose the need for further exploration. ML has demonstrated a potential to complement SCS to an extent ranging from assistance with candidate selection to replacing invasive and costly aspects of the surgery. The clinical application of ML in SCS shows promise for improving patient outcomes, reducing costs of treatment, limiting invasiveness, and resulting in a better quality of life for the patient.

Role of patient selection and trial stimulation for spinal cord stimulation therapy for chronic non-cancer pain: a comprehensive narrative review

Background/importance

Patient selection for spinal cord stimulation (SCS) therapy is crucial and is traditionally performed with clinical selection followed by a screening trial. The factors influencing patient selection and the importance of trialing have not been systematically evaluated.

Objective

We report a narrative review conducted to synthesize evidence regarding patient selection and the role of SCS trials.

Evidence review

Medline, EMBASE and Cochrane databases were searched for reports (any design) of SCS in adult patients, from their inception until March 30, 2022. Study selection and data extraction were carried out using DistillerSR. Data were organized into tables and narrative summaries, categorized by study design. Importance of patient variables and trialing was considered by looking at their influence on the long-term therapy success.

Findings

Among 7321 citations, 201 reports consisting of 60 systematic reviews, 36 randomized controlled trials (RCTs), 41 observational studies (OSs), 51 registry-based reports, and 13 case reports on complications during trialing were included. Based on RCTs and OSs, the median trial success rate was 72% and 82%, and therapy success was 65% and 61% at 12 months, respectively. Although several psychological and non-psychological determinants have been investigated, studies do not report a consistent approach to patient selection. Among psychological factors, untreated depression was associated with poor long-term outcomes, but the effect of others was inconsistent. Most RCTs except for chronic angina involved trialing and only one RCT compared patient selection with or without trial. The median (range) trial duration was 10 (0–30) and 7 (0–56) days among RCTs and OSs, respectively.

Conclusions

Due to lack of a consistent approach to identify responders for SCS therapy, trialing complements patient selection to exclude patients who do not find the therapy helpful and/or intolerant of the SCS system. However, more rigorous and large studies are necessary to better evaluate its role.

Spinal Cord Stimulation Meets Them All: An Effective Treatment for Different Pain Conditions. Our Experience and Literature Review

IntroductionSpinal Cord Stimulation (SCS) is an emerging minimally invasive technique which uses neuromodulation to manage different forms of intractable pain. SCS is a well-established option for the treatment of various pain conditions, and nowadays, indications are ever increasing.Materials and MethodsIn this study, we present our case series of 49 patients who underwent SCS at our Institution for the treatment of pain from different etiologies, and discuss our 10-year experience in SCS. For the purpose of this study, we also performed a systematic review of current indications and new perspectives in SCS.ResultsAmong our case series, patients were differentiated into two groups upon prior spinal surgery: patients who had undergone prior spinal surgery for back pain were defined as the "FBSS (failed back surgery syndrome) group," instead patient suffering from different types of pain but who had never undergone surgery were defined as the "naive group." As regards clinical response to SCS, 20 patients out of 36 (55.56%) were classified as responders in the FBSS group; in the "naïve" group, 10 patients out of 13 (76.92%) were classified as responders. Among the "not responders" group, several patients suffered from infections.Of the recent literature about SCS, 2124 records were screened and 37 studies were finally included in the qualitative synthesis for our systematic review.DiscussionIn case of FBSS, surgical revision is often associated with a high morbidity and corresponding low rates of success. Unfortunately, patients affected by chronic pain often become refractory to conservative treatments. Spinal Cord Stimulation (SCS) is nowadays considered as an effective therapy for several chronic and neuropathic pain conditions, such as failed back surgery syndrome. As regards the economic impact of SCS, implantation of an SCS system results in short-term costs increase, but the annual cumulative costs decrease during the following years after implantation, when compared to the costs of conventional management. Beyond the application for the treatment of FBSS, SCS has also been used for the treatment of other types of chronic non-oncological pain such as neuropathic pain and chronic back pain ineligible for surgical intervention. This evidence paved the way to establishing the potential role of SCS also for the treatment of oncological pain. However, the effectiveness and relative safety of SCS for cancer-related pain has not yet been adequately established.ConclusionsSpinal Cord Stimulation is a well-established treatment option in for FBSS. Beyond that, SCS has also been used for the treatment of "naive" patients, suffering from other types of chronic, both oncological and non-oncological, medical-refractory pain such as neuropathic pain and chronic back pain ineligible for surgical intervention.

Development of Machine Learning-Based Models to Predict Treatment Response to Spinal Cord Stimulation

BACKGROUND

Despite spinal cord stimulation's (SCS) proven efficacy, failure rates are high with no clear understanding of which patients benefit long term. Currently, patient selection for SCS is based on the subjective experience of the implanting physician.

OBJECTIVE

To develop machine learning (ML)-based predictive models of long-term SCS response.

METHODS

A combined unsupervised (clustering) and supervised (classification) ML technique was applied on a prospectively collected cohort of 151 patients, which included 31 features. Clusters identified using unsupervised K-means clustering were fitted with individualized predictive models of logistic regression, random forest, and XGBoost.

RESULTS

Two distinct clusters were found, and patients in the cohorts significantly differed in age, duration of chronic pain, preoperative numeric rating scale, and preoperative pain catastrophizing scale scores. Using the 10 most influential features, logistic regression predictive models with a nested cross-validation demonstrated the highest overall performance with the area under the curve of 0.757 and 0.708 for each respective cluster.

CONCLUSION

This combined unsupervised-supervised learning approach yielded high predictive performance, suggesting that advanced ML-derived approaches have potential to be used as a functional clinical tool to improve long-term SCS outcomes. Further studies are needed for optimization and external validation of these models.

Machine Learning Algorithms Provide Greater Prediction of Response to SCS Than Lead Screening Trial: A Predictive AI-Based Multicenter Study

Persistent pain after spinal surgery can be successfully addressed by spinal cord stimulation (SCS). International guidelines strongly recommend that a lead trial be performed before any permanent implantation. Recent clinical data highlight some major limitations of this approach. First, it appears that patient outco mes, with or without lead trial, are similar. In contrast, during trialing, infection rate drops drastically within time and can compromise the therapy. Using composite pain assessment experience and previous research, we hypothesized that machine learning models could be robust screening tools and reliable predictors of long-term SCS efficacy. We developed several algorithms including logistic regression, regularized logistic regression (RLR), naive Bayes classifier, artificial neural networks, random forest and gradient-boosted trees to test this hypothesis and to perform internal and external validations, the objective being to confront model predictions with lead trial results using a 1-year composite outcome from 103 patients. While almost all models have demonstrated superiority on lead trialing, the RLR model appears to represent the best compromise between complexity and interpretability in the prediction of SCS efficacy. These results underscore the need to use AI-based predictive medicine, as a synergistic mathematical approach, aimed at helping implanters to optimize their clinical choices on daily practice.