Deep Learning Algorithm Trained on Lumbar Magnetic Resonance Imaging to Predict Outcomes of Transforaminal Epidural Steroid Injection for Chronic Lumbosacral Radicular Pain

BACKGROUND

Transforaminal epidural steroid injections (TFESI) are widely used to alleviate lumbosacral radicular pain. Knowledge of the therapeutic outcomes of TFESI allows clinicians to elucidate therapeutic plans for managing lumbosacral radicular pain. Deep learning (DL) can outperform traditional machine learning techniques and learn from unstructured and perceptual data. A convolutional neural network (CNN) is a representative DL model.

OBJECTIVES

We developed and investigated the accuracy of a CNN model for predicting therapeutic outcomes after TFESI for controlling chronic lumbosacral radicular pain using T2-weighted sagittal lumbar spine magnetic resonance (MR) images as input data.

STUDY DESIGN

Imaging study using DL.

SETTING

At the spine center of a university hospital.

METHODS

We collected whole T2-weighted sagittal lumbar spine MR images from 503 patients with chronic lumbosacral radicular pain due to a herniated lumbar disc (HLD) and spinal stenosis. A "good outcome" was defined as a >= 50% reduction in the numeric rating scale (NRS-11) score at 2 months after TFESI vs the pretreatment NRS-11 score. A "poor outcome" was defined as a < 50% decrease in the NRS-11 score at 2 months after TFESI vs pretreatment.

RESULTS

In the prediction of therapeutic outcomes after TFESI on the validation dataset, the area under the curve was 0.827.

LIMITATIONS

Our study was limited in that we used a small amount of lumbar spine MR imaging data to train the CNN model.

CONCLUSIONS

We demonstrated that a CNN model trained, using whole lumbar spine sagittal T2-weighted MR images, could help determine outcomes after TFESI in patients with chronic lumbosacral radicular pain due to an HLD or spinal stenosis.

Comparison of the Effect of Balloon Catheter vs Nucleoplasty vs Balloon Catheter and Nucleoplasty in Patients With Lumbar Spinal Stenosis

BACKGROUND

Nucleoplasty and neuroplasty are often performed in patients with refractory lower back pain when conservative treatment is ineffective. Lumbar spinal stenosis (LSS) is caused by multiple factors; in some cases, a single procedure of nucleoplasty or neuroplasty alone does not provide sufficient treatment effect.

OBJECTIVES

This study aimed to investigate and compare the pain relief and pain-free interval among patients with LSS who underwent nucleoplasty, neuroplasty, and combined balloon neuroplasty and nucleoplasty.

STUDY DESIGN

Retrospective study.

SETTING

In-ha University hospital pain clinic.

METHODS

This is a retrospective study of the medical records and survey of 98 patients with LSS who visited a pain clinic between 2019 and 2020 and underwent nucleoplasty, neuroplasty, and combined balloon neuroplasty and nucleoplasty. Patients with disc height &lt; 50% of the adjacent disc on magnetic resonance imaging and those with moderate and severe extraforaminal stenosis were excluded. Thus, 60 patients who underwent nucleoplasty (n = 20), neuroplasty (n = 20), and combined balloon neuroplasty and nucleoplasty (n = 20) for LSS were analyzed. The patients were instructed to rate their pain intensity via an 11-point numeric rating score (NRS) before and after the procedure. The Korean version of the Oswestry Disability Index (ODI) questionnaire was checked before and after the procedure.

RESULTS

The pain intensity decreased to NRS 3 ± 0.14 and 1.85 ± 0.19 in the nucleoplasty and combined balloon neuroplasty and nucleoplasty groups, respectively, indicating a significant difference (P = 0.003). ODI was significantly decreased after the procedure compared with that before the procedure in all groups. After the procedure, ODI decreased to 13.89 ± 0.20 and 11.21 ± 0.33 in the nucleoplasty and combined balloon neuroplasty and nucleoplasty groups, respectively, with a significant difference between the 2 groups (P &lt; 0.05). The patients in the nucleoplasty group achieved pain relief for 4.93 ± 1.22 months after the procedure, whereas those in the balloon neuroplasty group achieved pain relief for 5 ± 1.37 months. In the combined balloon neuroplasty and nucleoplasty group, pain relief was maintained for 10.2 ± 1.11 months (P = 0.003).

LIMITATIONS

The pain was assessed with NRS without considering the patients' pain medication. There may be differences in the outcome of the procedure depending on the surgeon.

CONCLUSION

The pain reduction effect was greater and was retained for a longer period with combined balloon neuroplasty and nucleoplasty than with nucleoplasty or neuroplasty alone.

Predicting Epidural Space Spread Using Ultrasound Color Doppler Imaging in Interlaminar Epidural Steroid Injection: A Prospective Observational Study

BACKGROUND

While the use of fluoroscopy-guided transforaminal epidural steroid injection (TFESI) to help spread the injectate toward the ventral side has increased, this procedure has a radiation risk. Recently, ultrasound has been widely used in the medical field; among ultrasound methods, color Doppler is useful for predicting the direction of the injectate.

OBJECTIVE

This study describes a novel technique employing color Doppler to help predict epidural space spread in interlaminar epidural steroid injection (ILESI).

STUDY DESIGN

Prospective observational study.

SETTING

The study took place at a single pain clinic within a medical center in Jeonju, Republic of Korea.

METHODS

We enrolled 35 patients scheduled for lumbar epidural steroid injection (ESI).  Ultrasound-guided epidural lateral parasagittal interlaminar injection was performed and real-time images using color Doppler were recorded during injections of 5 mL of 0.1% ropivacaine containing contrast dye with dexamethasone 5 mg (1 mL). Fluoroscopy-guided TFESI was performed if it was difficult to perform the procedure based on ultrasound images.

RESULTS

The analysis included 30 images from 30 patients. The observed sensitivity, specificity, positive predictive value, and negative predictive values of the ultrasound color Doppler were 100%, 89.5%, 84.6%, and 100%, respectively. The agreement with ultrasound color Doppler was 93.3%.

LIMITATIONS

The sample size was relatively small.

CONCLUSION

The main advantage of ultrasound-guided ILESI is the lack of radiation exposure and contrast medium requirement. Color Doppler may be a reliable imaging modality to predict epidural space spread during ultrasound-guided ILESI. It is worth predicting the spread in the anterior epidural space (AES) by first attempting ultrasound-guided ESI. If the injectate has not spread to the AES, fluoroscopy-guided TFESI may be a good option after confirming improvement of the patient's symptoms.

Declining Utilization Patterns of Percutaneous Adhesiolysis Procedures in the Fee-For-Service (FFS) Medicare Population

BACKGROUND

Percutaneous epidural adhesiolysis is a minimally invasive therapeutic modality used in the treatment of patients with chronic low back and lower extremity pain, often recalcitrant to other modalities including epidural injections and surgical interventions. While the initial utilization since its introduction and development of appropriate Current Procedural Terminology (CPT) codes increased up until 2008, but since 2009, there has been a significant decline in utilization of these procedures in the Medicare population. These procedures declined by 53.2% at an annual rate of 10.3% from 2009 to 2016. A recent update analysis on the reversal and decline of growth of utilization of interventional techniques in managing chronic pain in the Medicare population from 2009 to 2018 revealed an even further decline of adhesiolysis procedures.

STUDY DESIGN

An analysis of the utilization patterns of percutaneous adhesiolysis procedures in managing chronic low back and lower extremity pain in the Medicare population from 2000 to 2018, with comparative analysis from 2000 to 2009 and 2009 to 2018.

OBJECTIVE

To assess the utilization patterns of percutaneous adhesiolysis in managing chronic low back pain in the Medicare population.

METHODS

The Centers for Medicare and Medicaid Services (CMS) Physician Supplier Procedure Summary Master of Fee-For-Service (FFS) Data from 2000 to 2018 was used.In this analysis, various variables were assessed in reference to usage patterns of percutaneous adhesiolysis procedures with analysis of growth or declining utilization patterns. We also assessed specialty-based utilization, as well as statewide utilization.

RESULTS

The decline of percutaneous adhesiolysis procedures began in 2009 and has continued since then. From 2009 to 2018, the overall decline was 69.2%, with an annual decline of 12.3% compared to an overall 62.6% increase from 2000 to 2009, with an annual increase of 5.6%. Compared to multiple other interventions, including epidural injections and facet joint interventions, percutaneous adhesiolysis has declined at a rapid rate.

CONCLUSIONS

This assessment in the FFS Medicare population in the United States shows an irreversible decline of utilization of percutaneous adhesiolysis procedures, which has been gradually deteriorating with a 69.2% decline from 2009 to 2018 with an annual decline of 12.3% during that same time period.