Article Text
PDF
Abstract
Background Nusinersen is the only approved treatment for all spinal muscular atrophy (SMA) subtypes and is delivered intrathecally. Distorted spinal anatomy and instrumentation preclude standard approaches for intrathecal access, necessitating alternative techniques for delivery. The purpose of this study is to report technical success and adverse events of transforaminal intrathecal delivery of nusinersen.
Methods 28 patients, mean age 24.1±9.8 years (range 10.0–51.0 years), with intermediate or late onset SMA, underwent a combined 200 transforaminal nusinersen injections. All patients had osseous fusion or spinal instrumentation precluding standard posterior access routes. Patients who underwent nusinersen injections using a technique other than transforaminal lumbar puncture (n=113) were excluded. Technical success, adverse events (AEs) and radiation exposure were recorded.
Results 200 (100%) procedures were technically successful; 6 (3%) required a second level of attempt for access. 187 (93.5%) interventions were completed using cone beam computed tomography (CBCT) with two-axis fluoroscopic navigational overlay. 13 (6.5%) procedures were performed with fluoroscopic-guidance only at subsequent sessions. There were 8 (4.0%) mild AEs and 2 (0.5%) severe AEs; one patient received antibiotics for possible traversal of the large bowel but did not develop meningitis, and one patient developed aseptic meningitis. Mean air kerma was 74.5±161.3 mGy (range 5.2–1693.0 mGy).
Conclusion Transforaminal intrathecal delivery of nusinersen is feasible and safe for gaining access in patients with distorted spinal anatomy. The use of CBCT delineates anatomy and optimizes needle trajectory during the initial encounter, and may be used selectively for subsequent procedures.
- CT
- intervention
- navigation
- pediatrics
- spine
Statistics from Altmetric.com
Introduction
Spinal muscular atrophy (SMA) is an inherited neuromuscular disorder that causes progressive loss of motor neurons in the anterior horn cells resulting in profound neuromuscular weakness.1 Intrathecal nusinersen (Spinraza; Biogen Inc, Boston, MA) is the only treatment approved by the Food and Drug Administration for all SMA subtypes.2 Scoliosis frequently develops in patients with intermediate and late onset SMA (types II–IV) secondary to neuromuscular weakness and often requires spinal instrumentation to prevent worsening deformities.1 Intrathecal drug delivery may be challenging and traditional lumbar puncture techniques are often inadequate for achieving intrathecal access.3 Patients with severe or early onset SMA (type I, Werdnig-Hoffman disease) typically die before the age of 2 years without ventilatory support and do not live long enough to develop severe spinal deformities.4
Early results of transforaminal intrathecal injections of nusinersen have been reported.5–9 The purpose of this study was to report technical success and adverse events (AEs) of transforaminal intrathecal delivery of nusinersen for patients with intermediate and late onset SMA using either cone beam computed tomography (CBCT) with two-axis fluoroscopic navigational overlay or fluoroscopic-guidance only.
Methods
Study design
This study was conducted with Institutional Review Board approval and complied with the Health Insurance Portability and Accountability Act. This study was assessed using the Strengthening the Reporting of Observational studies in Epidemiology (STROBE) guidelines.10
Inclusion and exclusion criteria
Between September 2017 and December 2019 (28 months), 28 patients with intermediate or late onset SMA received nusinersen delivered intrathecally via a transforaminal approach. All patients included in this study were referred for interventional radiology (IR) after prior failures of drug delivery or if traditional lumbar puncture techniques were deemed infeasible given complex spinal anatomy. Patients 1–4 and injections 1–15 of this study comprised the early study on technical results and safety previously reported.5 Transforaminal nusinersen delivery was achieved either using CBCT with two-axis fluoroscopic navigational overlay or fluoroscopic guidance only following prior access using CBCT. Patients who underwent nusinersen injection by IR using any technique other than transforaminal lumbar puncture were excluded (n=113).
Patient demographics
Patient demographics are shown in table 1. Twenty-eight consecutive patients were included in the study: 26 (92.9%) patients had SMA type II, and two (7.1%) patients had SMA type III disease. Mean patient age was 24.1±9.8 years (range 10.0–51.0 years). All patients had distorted spinal anatomy (eg, severe scoliosis, osteophyte formation) and posterior osseous fusion precluding standard access routes. Twenty-five (89.3%) patients had prior surgical instrumentation with fusion while three (10.7%) patients had spontaneous osseous fusion. For patients ≤18 years of age at the time of referral, all but one had undergone prior posterior fusion. A single pediatric patient had spontaneous osseous fusion and a history of failed posterior approach lumbar puncture. All patients, regardless of age, were referred based on their spinal anatomy, surgical history and/or imaging with anticipated technical challenges for access and medication delivery.
Patient demographics
Techniques
Intrathecal nusinersen injections using a transforaminal approach with CBCT have been previously described.5 General anesthesia, moderate sedation, and local anesthesia were used in 94 (47.0%), 35 (17.5%), and 71 (35.5%) procedures, respectively. One hundred and ninety-nine (99.5%) procedures were performed with a 22-gauge beveled spinal needle (Quincke, Becton Dickinson, Franklin Lakes, NJ), and a single (0.5%) case was performed with a pencil-point 24-gauge spinal needle (Sprotte, Teleflex Medical Company, Limerick, PA).
With cone beam navigation
The cone beam navigation technique is shown in figure 1. Patients were positioned in the lateral decubitus position with the convex curvatures of the scoliotic spine facing upwards to expose the foramen of interest. CBCT guidance was performed using a Philips angiography system and proprietary XperGuide navigation (Philips; Eindhoven, the Netherlands). A non-contrast CBCT of the z-axis region of interest as determined by the preprocedural imaging was acquired. The intrathecal target site and skin access point were identified and selected using the XperGuide software. Local anesthesia was applied at the skin entry site and a spinal needle was advanced, under intermittent fluoroscopic guidance with orthogonal views, through the neural foramen and into the intrathecal space. Position was confirmed by return of cerebrospinal fluid after stylet removal; following removal of 5 mL of cerebrospinal fluid, 12 mg of nusinersen was delivered in 5 mL total volume over 2 min.
Patient with type II spinal muscular atrophy and prior spinal fusion. Axial (A) and sagittal (B) cone beam CT images demonstrate conspicuity of the exiting nerve root (arrows) despite metal artefact. A needle route is planned using XperGuide navigation (Philips; Eindhoven, the Netherlands), passing along the posterior crescent (C, D) to avoid the nerve root during access.
Without cone beam navigation
Procedures performed solely with fluoroscopic guidance without the aid of CBCT were only undertaken for patients who had previously undergone an injection with CBCT. Patients were selected to undergo injections with fluoroscopic-guidance only in cases where prior CBCT-guided procedures had delineated the anatomy and optimal needle trajectory in a manner that was felt to be replicable with fluoroscopic guidance alone. In these cases, patients were placed in the lateral decubitus position. The neural foramen of interest was identified using fluoroscopy and, after administering local anesthesia, a spinal needle was advanced under intermittent fluoroscopic guidance into the intrathecal space. The procedure was then completed in the same fashion as for CBCT-guided injections.
Instillation practices and clinical evaluation
Injections occurred in three loading doses separated by 2 weeks, a fourth dose a month after the third, and indefinite maintenance doses every 4 months afterwards.11 Patients were monitored for immediate (during the procedure) complications before discharge and questioned about delayed (within 30 days) complications, including neuropathic pain or post-lumbar puncture headache, when they returned for repeat injections. Outpatient clinic notes were reviewed for identification of potential treatment-related complications.
Variables collected and defined
Primary outcome was technical success and secondary outcomes included procedural AEs, fluoroscopy time, radiation dose, and clinical follow-up. Technical success was defined as gaining access to the thecal sac and complete delivery of the 12 mg dose, inclusive of multiple access attempts during a single procedural encounter. Primary technical success was defined as successful access via the initially planned foramen. Secondary success was defined as requiring an additional attempt at another spinal level during the same procedural encounter. Failure was defined as any other outcome. Complications were recorded according to the Society of Interventional Radiology (SIR) guidelines on AEs.12 Fluoroscopy time was defined as the total time that fluoroscopy was used during the intervention and was recorded in 0.1 min increments. Air kerma was defined as the air kerma accumulated in space relative to the fluoroscopic gantry and recorded in mGy. Follow-up time was defined as the number of days elapsed between the procedure and the time of data collection.
Statistical analyses
Calculations of means, standard deviations and ranges were performed with statistical spreadsheet software (Microsoft Excel; Microsoft; Redmond, WA).
Results
Technical successes
The results are shown in table 2. Two hundred (100%) transforaminal injections were successfully performed within the procedural encounter: 187 (93.5%) were performed using CBCT navigation and 13 (6.5%) were completed with fluoroscopic guidance only.
Technical successes, adverse events (AEs), fluoroscopy time, and air kerma.
One hundred and ninety-four (97.0%) cases were primarily successful. Six (3.0%) interventions were secondarily successful and required a second access approach, at a different spinal level, during the same procedural encounter. Three (1.5%) instances were due to technical difficulties in gaining intrathecal access via the initially planned route. In all six (3.0%) instances, CBCT was performed before the start of the procedure. In one (0.5%) of these instances the procedure was interrupted, and the patient was transported from the angiography suite to undergo a diagnostic CT evaluation after the needle tip was seen within the thecal sac on CBCT, but cerebral spinal fluid (CSF) could not be aspirated. This patient had not undergone a preprocedural planning CT. The patient returned, after the CT, with subsequent successful drug delivery following identification of an alternative needle trajectory. One (0.5%) procedure was moved to another spinal level due to patient discomfort, and one (0.5%) was moved after a prominent vein was inadvertently accessed.
Adverse events
Eight (4.0%) mild AEs (SIR class I, requiring no or nominal therapy) occurred. Two (1.0%) patients had transient intraprocedural radicular pain which resolved without intervention. Two (1.0%) patients reported self-limited spinal headaches. One (0.5%) patient had transient radicular pain which resolved after needle removal. One (0.5%) patient had headache and neck pain with radiation down to the back, persisting for 4 days and prompting return to the emergency department. The CT was negative for epidural hematoma. The patient was treated with gabapentin and pain had resolved before a clinic visit 1 week post-procedure. In one (0.5%) patient, the spinal needle returned blood without clearing. The needle was removed, and manual pressure was held until hemostasis was achieved. One (0.5%) patient had a neuropathic headache attributed to positioning during the procedure which resolved with the use of gabapentin. One patient’s family requested the procedure to be performed with a 24-gauge pencil-point spinal needle on presentation for the patient’s sixth procedure. There were no prior spinal headaches and the request was made to reduce the chance of future spinal headache. A single procedure performed with this needle was uncomplicated.
Two (0.5%) severe AEs (SIR class III, requiring marked escalation of care) occurred. One (0.5%) patient presented with symptoms concerning for meningitis 4 days after nusinersen injection and was admitted for further evaluation. CSF cultures were negative and the patient was diagnosed with aseptic or chemical meningitis; this was attributed to nusinersen injection, given reports in post-marketing surveillance.13 The patient was discharged after receiving supportive care. In the other severe AE, the procedure was initially attempted with fluoroscopic guidance only, on a side not previously accessed in this patient. CBCT was obtained after poor return of CSF. CBCT showed the needle tip to be in the central canal, but also raised concern for possible traversal of large bowel. The procedure was subsequently successfully performed via a new access trajectory. The patient was observed and treated with 48 hours of intravenous ampicillin, cefepime and metronidazole at meningitic dosing. CSF cultures were negative, antibiotics were stopped, and the patient was discharged without any adverse sequelae.
Radiation dose
Mean fluoroscopy time was 1.7±1.6 min (range 0.1–11.0 min). Mean air kerma was 74.5±161.3 mGy (range 5.2–1693.0 mGy).
Clinical follow-up evaluation
Patients were monitored for immediate (during the procedure) complications before discharge, and were questioned about delayed (within 30 days) complications including neuropathic pain or post-lumbar puncture headache when they returned for repeat injections. Mean clinical follow-up was 430.4±248.0 days (range 25.0–975.0 days).
Discussion
The efficacy of nusinersen is established for the treatment of SMA for both early and late onset disease.14 15 As nusinersen currently remains the only treatment available for those above the age of 2 years, referrals for intrathecal drug delivery are expected to increase for both pediatric and adult patients with complex spine anatomy.
The efficacious and safe use of a transforaminal approach using CBCT and two-axis fluoroscopic navigational overlay for pediatric patients with SMA and extensive spinal instrumentation precluding standard lumbar puncture techniques has been previously described.5 Multiple groups subsequently published experiences with transforaminal injection of nusinersen with high technical success rates and minimal-to-no associated complications.6–9 This represents the largest known series to date, with 200 total procedures using the transforaminal approach for the delivery of nusinersen, and similarly indicates that this technique is reliably safe and successful.
Only minor AEs occurred in instances where CBCT was utilized. Minor AEs included self-resolving spinal headaches and transient radicular pain. One patient continues to experience radicular pain during serial follow-up visits. Multiple studies have helped elucidate the neurovascular anatomy of the intervertebral foramen.16 17 Knowledge of the most likely anatomic location of the spinal nerve roots and vascular pedicles in the intervertebral foramen may aid in minimizing procedural AEs. The nerve root often can be visualized on CBCT and a needle trajectory may then be planned to avoid it. When the exiting nerve route cannot be visualized, the middle of the posterior crescentic segment of the intervertebral foramen should be empirically targeted to avoid neurovascular structures. The anterior-superior quadrant of the intervertebral foramen should be avoided as radicular arteries frequently occupy this location.16
One major AE involved possible needle traversal of large bowel en route to the intrathecal sac. This occurred in a procedure initially relying on fluoroscopic guidance without CBCT and the complication was discovered after CBCT was obtained. Ultimately, the patient suffered no adverse sequelae and CSF cultures did not demonstrate any evidence of infectious meningitis. While the patient had undergone successful CBCT-guided treatment previously, that access route was not replicated during the complicated attempt. While transforaminal access may be achieved safely with fluoroscopy alone, the authors recommend using safe access routes as determined by antecedent CT or CBCT. The other major AE, aseptic meningitis secondary to nusinersen injection, did not have any long-term sequelae for the patient. However, it is important to be aware of this potential complication and discuss it with patients as part of the informed consent process. A minority of the procedures in this series were performed with local anesthesia alone, largely attributable to the inclusion of pediatric patients and higher rates of sedation in this population. Although no sedation-related complications occurred, the fragility of patients with SMA and associated anesthetic risks should not be underestimated. While procedures ideally proceed with minimal or no sedation when possible, an individualized treatment plan addressing patient comfort (and stillness during navigational overlay guidance) results in technical success.
As the number of patients receiving nusinersen increases, it will be important to implement a reliable and safe system for the intrathecal administration of this drug in a patient population. Other approaches have been described, including cervical spine puncture and surgical laminotomy, for the administration of nusinersen.18 19 While no major complications were reported in the aforementioned studies, these approaches are more invasive and likely carry a higher procedural risk. A novel subcutaneous intrathecal catheter system with an implantable fusion port has also been used with success in 10 patients.20 While the initial procedure to place the subcutaneous intrathecal catheter is more invasive than CBCT-guided transforaminal lumbar puncture, subsequent drug deliveries are expeditious and well tolerated.
Limitations of this study include a lack of standardized follow-up or a questionnaire specific to complication surveillance. The assessment of complications was reliant on patient reporting. Intraprocedural discomfort was not objectively quantified and may be inconsistently reported in the electronic medical record. Additionally, acute intraprocedural radicular pain may be masked by anesthesia. Motor nerve injuries may also be potentially masked by pre-existing and underlying neurologic deficits in this patient population.
Conclusion
Transforaminal intrathecal delivery of nusinersen is reliable and safe for gaining intrathecal access in patients with distorted spinal anatomy and hardware precluding traditional lumbar puncture technique. This technique may be accomplished using CBCT with two-axis fluoroscopic navigational overlay or with fluoroscopic-guidance only. The use of CBCT delineates anatomy and optimizes needle trajectory during the initial encounter and may be used selectively for subsequent procedures.
Footnotes
Twitter @Eric_Monroe_MD
Contributors All authors met the criteria for authorship. JJW, JFBC, and EJM analyzed and interpreted the data, and wrote the article, and thereby take responsibility for this work. DKH, JFBC and EJM conceived the study and participated in its design and coordination. SV, DSS, NN, NR, JR, KSHK, GMS, AT, SA assisted with manuscript composition. Conception and design of this work were discussed with all of the authors. All authors agreed to publish this work and critically reviewed the article.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests EJM discloses scientific advisory and speaking for Biogen, Inc. NN receives research support for studies for Biogen, Inc.
Patient consent for publication Parental/guardian consent obtained.
Ethics approval All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. The study did not receive funding. Written consent was obtained for anonymized pictures displayed in Figure 1. Informed consent for data was waived from all individual participants included in the study. The study was approved by the institutional ethics committee at Seattle Children’s Hospital (institutional registration numbers: IRB00000277, IRB00009311; study number: STUDY00002336).
Provenance and peer review Not commissioned; externally peer reviewed.
Data availability statement All data relevant to the study are included in the article or uploaded as supplementary information.