Foreword

Epidural spinal cord stimulation (SCS) provides substantial potential for improving the health and quality of the life of individuals with spinal cord injury (SCI). With mounting scientific evidence in support of this intervention, the National Institute of Biomedical Imaging and Bioengineering (NIBIB) explored and assessed the most current information about SCS for individuals with SCI. We sought the collaboration by colleagues at NINDS, NICHD, and the FDA. In addition, we received the cooperation and input from key scientists and researchers in this new and emerging field. Our goals were to encourage an informed discussion of the uses of SCS, to identify all pertinent issues in the study and application of this intervention, and create an organized collaboration among research stakeholders to advance the availability and use of SCS. As a result, a Consortium was created which facilitated directed communication among the scientists, and resulted in a framework for studying the effectiveness of SCS. The Consortium members conferred on a frequent basis over a period of eight months to identify the various issues which must be addressed in order for this promising new field to advance and determine the efficacy of SCS in a broader population. These issues include the scientific, medical, economic, and regulatory implications of this new approach for individuals with SCI. In the Report below, a framework is presented for examining SCS in its application to people with SCI. The therapeutic goal is improving key autonomic functions - bladder, bowel, and sexual - effectively and safely.

We welcome your comments.  Please send these by September 30, 2015 to info@nibib.nih.gov.

Roderic I. Pettigrew
Director, NIBIB

Roderic I. Pettigrew, Ph.D, M.D.
Director
National Institute of Biomedical Imaging and Bioengineering
National Institutes of Health
Building 31, Room 1C14
Bethesda, MD  20892-2281
301-496-8859
301-480-0679 - FAX
rpettigrew@nih.gov
http://www.nibib.nih.gov 
 

Introduction

Individuals with spinal cord injury (SCI) face devastating health problems that have a dramatic impact on their lives and the lives of their family members. Despite all the public health efforts to prevent SCI, the annual numbers of individuals have not diminished in the last decade and represent a major challenge to not only the health care system but also to those whose lives have been dramatically altered by SCI. Although the incidence rate of SCI has remained relatively stable during the last decade (53 to 54 cases per 1 million population), there has been a modest increase in the overall absolute number of cases of SCI from 1993 to 2012 (Jain et al. (2015)). In 1993 the prevalence increased from an estimated 13,706 to an estimated prevalence of 16,965 in 2012. In 2009, the ratio of males to females was 4.2/1, or 80.9% of SCI occurring among males (National Spinal Cord Injury Statistical Center (NSCISC), 2009). In a recent state of Oklahoma study on SCI by age, incidence rates of spinal cord injury were highest in the male 20- to 24-year-old population (annual rate, 144 cases/million) followed by the male 15- to 19-year-old population. It is interesting to note that spinal cord injury incidence rates have substantially increased over time in the 65- to 74-year-old and 75- to 84-year-old age groups. This now accounts for a larger proportion of total spinal cord injury cases because of the increasing elderly population (Jain et al., 2015).

Recent major successes in the use of epidural spinal stimulation in people with spinal cord injury (SCI) have generated a pressing need to determine a course of action for accelerating research in this area, with the ultimate goal of making available a safe and effective therapy to individuals with SCI in a timely manner (Angeli, 2014; Capogrosso, 2013; Edwards, 2013; Ellaway, 2007; Herman et al., 2002; Gad, 2014; Harkema, 2011).  In response to this need, the National Institute of Biomedical Imaging and Bioengineering (NIBIB)—one of several federal funders of this research within the National Institutes of Health (NIH)—initiated and led a series of meetings from November 2014 to June 2015 with key academic, industry, patient advocacy, and medical leaders. The goals of the meetings were: (1) to provide a forum for information exchange, (2) to identify the next steps required to accelerate this research, and (3) ultimately increase the availability of this new intervention to those whom it might benefit. A major question posed throughout the meetings was how to devise a prospective research study of epidural spinal stimulation that addresses problems of high-relevance to the SCI community. We also sought interventions that if successful, would have an immediate impact on the lives of individuals with SCI, and could be made more widely available in five years.    

While the recent breakthrough research studies in epidural spinal stimulation sought improved motor control as a primary goal, individual reports from each of these studies suggest that epidural stimulation may also improve the lives of individuals with SCI in other meaningful ways, such as alleviating autonomic dysfunctions and secondary complications. The ability to generate complex and coordinated locomotor movement using epidural spinal stimulation in individuals with SCI is an extremely challenging long-term goal that requires extensive research time and effort on both the part of individuals and researchers. Thus, it was realized early in discussions that a study focusing on alleviating autonomic dysfunction could potentially provide a more feasible initial step for determining safety and efficacy of epidural spinal stimulation, using existing devices in the population of individuals with SCI. Bladder dysfunction was chosen as an initial focus of this endeavor because it has been identified as one of the most significant factors that negatively impact quality of life for people with SCI (Anderson, 2004; Ditunno, 2008; Liu, 2009; Ethans, 2014). In addition to limiting an individual’s autonomy and ability to engage socially, bladder dysfunction and complications associated with chronic and intermittent catheterization frequently leads to serious clinical conditions, including urinary tract infections, urinary incontinence, calculi, bladder cancer, and renal dysfunction. Such adverse outcomes result in hospital admissions and increase the risk of premature death (Anderson, 2004; Pagliacci, 2007; Hasma et al., 2007; Cardenas et al., 2004).

Electrical stimulation to improve bladder function has been attempted in people with SCI with varying success. In clinical practice the Brindley bladder control system was designed to provide sacral anterior root stimulation which allowed bowel or bladder emptying on demand in patients with SCI. More than 3,500 individuals worldwide received the implant. This technique unfortunately required a sacral rhizotomy (surgical procedure that interrupts the reflex arcs to denervate the bladder from its afferent and/or efferent pathways, thereby suppressing detrusor overactivity and autonomic dysreflexia in patients with spinal cord injury) which was permanent and the number of patients accepting the device dwindled over the years. As a result, this device is no longer manufactured in the U.S.; however, in those patients who did receive the implant, many achieved long-term success (Martens et al., 2011; Cho et al., 2012). 

Another technique, sacral neuromodulation, has been extensively utilized to treat urgency incontinence and non-obstructive urinary retention in patients without SCI since 1997.  The technique requires only percutaneous access to the sacral foramina making it minimally morbid and relatively simple to perform. Unfortunately, the success in SCI has been poor with this type of stimulation to a single afferent nerve and only individuals with partial SCI have seen any benefit (McGee et al., 2015).

In the recent discussions, it was agreed that bowel and sexual dysfunction should be studied in parallel, as the neural control of these systems also appear to be affected by epidural stimulation aimed at neural areas that modulate bladder function. Moreover, as discussions progressed, it became clear that prior to initiation of a large pivotal trial, one or more smaller pilot studies were needed to better characterize the epidural spinal stimulation parameters required to effectively improve bladder, bowel, and sexual function in individuals with SCI. This paper provides recommendations from a broad group of stakeholders for consideration in designing such a pilot study.

Approach

NIBIB convened an initial group of academic, industry, patient advocacy, and medical leaders in a workshop entitled “Addressing Paralysis through Spinal Stimulation Technologies,” which took place on November 15, 2014. The workshop was an unprecedented effort to gather experts from a wide range of disciplines and fields for the purpose of determining the best path forward for epidural spinal stimulation research in SCI individuals and making an effective treatment for paralysis available to more people it might benefit.\

In preparation for the workshop, participants received copies of scientific publications pertinent to the topic (Carhart et al., 2004; Huang et al., 2006; Ellaway et al., 2007; Harkema et al., 2011; Edwards et al., 2013; Angeli et al., 2014; Calancie and Alexeeva, 2014; Gad et al., 2014). The workshop began by discussing current epidural spinal stimulation technologies, outcomes from recent animal and human studies, and what is currently known about the mechanisms of action. Later in the meeting, participants were asked to prioritize conditions in individuals with SCI to be addressed and cohorts to be studied in future trials. They were also asked to define issues for consideration in determining stimulation parameters and outcome measures based on the current technology. 

Two key outcomes were realized from the workshop: (1) the formation of a consortium of academic, industry, patient advocacy, and medical leaders to facilitate greater communication and cooperation; and (2) the recognition that continued discussion and prioritization of conditions, cohorts, stimulation parameters, and clinical assessment measurements are needed in order to develop a roadmap for future research and treatment delivery to SCI patients. 

To continue the discussions which started at the workshop, a series of bi-weekly videoconferences were scheduled. Each videoconference focused on discussion topics that represented key issues related to the goal of determining the best approach for developing clinical studies for epidural spinal stimulation to improve bladder, bowel, and sexual function in individuals with SCI. The following schedule was implemented:

March 19, 2015: Topic: “Interventions, Study Objectives and Hypotheses”

April 2, 2015: Topic: “Clinical Testing and Training”

April 16, 2015: Topic: “Participant Selection and Length of Follow Up”

May 7, 2015: Topic: “End Points, Evaluation of Success, Safety and Anticipated Adverse Events”

June 4, 2015: Topic:  “Final Report of Consortium”

These bi-weekly videoconferences resulted in the identification of challenges related to methodology and precautions based on both animal and human studies.

Interventions, study objectives, and hypotheses

Consortium members described the initial intervention strategy as using epidural stimulation of the lumbar or sacral areas of the spinal cord (with electrodes covering the  spinal levels from L1 to S3) in participants with SCI. Details of the intervention require further optimization (Canby et al., 2014). With input from investigators who have had experience with implanted study participants, the Consortium emphasized the importance of conducting mapping studies, using rapid onset metrics of effective stimulation parameters and electrode locations for each study participant given that functional differences exist among participants with SCI. Industry representatives from the Consortium recommended that computational modeling should also be used to explore the theoretical effects of changing stimulation parameters of location, intensity, and duration. Although locomotor training and epidural stimulation combined have been noted to have positive impacts on autonomic function, there is a need to examine the impact of epidural spinal stimulation alone. To stay focused on the short-term opportunities, the Consortium concurred that the objectives for the initial study should relate to improving bladder, bowel, and sexual function collectively, as each have related and integrated neural pathways. The following  should be considered as potential outcome measures: increased bladder capacity, increased time between bladder emptying, increased ability to void voluntarily, reduction in urinary incontinence, reduced time needed for bowel management, reduction in fecal incontinence, improved sexual function, and improved health related quality of life and well-being, such as participation in the community, social activities, and employment.

As the precise mechanisms underlying spinal epidural stimulation are still unknown, discussions were prompted about the need for completing further animal studies before proceeding with human studies. Consortium members argued for pursuing both (1) additional animal studies to refine and develop hypotheses and general principles of function, and for (2) enabling immediate pilot studies with humans to identify relevant parameters and define the benefit and safety evidence in humans based on the promising results already observed. This approach parallels the approach taken with deep brain stimulation, in which the mechanism is also not well understood and requires further research; the impact in some clinical indications has been proven and in others is promising (Chiken & Nambu, 2015; Alamri et al., 2015). Notably the Consortium discussed both the benefits and limitations of the animal research and models. It was recognized that animal studies play a major role not only in revealing basic phenomena that may have clear clinical implications in possible strategies to recover  autonomic function, but they also play a continuing role in understanding the underlying mechanisms of these phenomena. The animal studies also help to provide insight into the strategies that can be used to optimize interventions to promote motor and autonomic function. Consequently they provide valuable information that can facilitate the translation of ideas to clinical human studies. Nonetheless, the Consortium emphasized that in some instances the efficacy of spinal stimulation technologies can best be tested by direct studies in humans.

In discussing possible hypotheses of function, the Consortium noted that studies directed toward using epidural stimulation and motor training have shown the recovery of the ability to stand and step with assistance and move the lower limbs voluntarily. Given the results observed in completely paralyzed individuals indicating improved bladder, bowel, and sexual function, it was hypothesized that the spinal and autonomic networks that generate the control of these three physiological systems can be reactivated to a more normal functional state via activity-dependent mechanisms. These include epidural stimulation of target networks and the associated autonomic ganglia in concert with motor training (standing and stepping). The Consortium identified mechanism questions that focused on (1) identifying the specific nerve fibers recruited by spinal cord stimulation and (2) where those fibers project in the central nervous system to produce subsequent activation of circuits that affect muscle tone in sphincters and bladder, bowel, sexual, and other autonomic functions.

Neuroplasticity of these spinal cord circuits should also be considered, as it is known that injury leads to changes in circuit function and chronic stimulation can also produce changes. Thus, understanding the direct and immediate effects of SCS along with longer-term training effects will be important.

The proposed hypotheses would be focused on the safety of the intervention and the voluntary control of bladder, bowel, and sexual function by the participants as a result of the intervention as follows:

Overall:

Hypothesis #1:  Spinal epidural stimulation is not associated with meaningful adverse events

Hypothesis #2:  Spinal epidural stimulation is not associated with increase in urinary, bowel, or sexual dysfunction

Hypothesis #3:  Use of epidural stimulation provides a measurable impact on health-related quality of life

Hypothesis #4:  Spinal epidural stimulation results in fewer episodes of autonomic dysreflexia related to bladder, bowel, or sexual activity

Bladder:

Hypothesis #5:  Bladder capacity is increased with appropriate epidural stimulation

Hypothesis #6:  Sphincter activity is increased during bladder filling with epidural stimulation

Hypothesis #7:  Appropriate epidural stimulation improves  coordination between the detrusor muscle  and sphincter muscle

Hypothesis #8:  The volume of residual urine is decreased after voiding with epidural stimulation relative to voiding without stimulation

Hypothesis #9:   The time it takes to completely void is reduced with epidural stimulation

Hypothesis #10:  Voiding with epidural stimulation reduces the dependence on clean intermittent catheterization, the incidence of upper and lower urinary tract infections, kidney damage and related hospitalization

Bowel:

Hypothesis #11:  Epidural stimulation decreases total time devoted to a bowel care program

Hypothesis #12:  Epidural stimulation decreases episodes of constipation and/ or incontinence

Hypothesis #13  Epidural stimulation improves the regularity of bowel movements

Hypothesis #14:  Epidural stimulation decreases the need for bowel medications, suppositories, and/ enemas

Hypothesis #15:  Epidural stimulation decreases the need for manual maneuvers performed digitally and/ or with assistive/adaptive devices

Hypothesis #16:  Anorectal manometry and/or defecography measures improve with epidural stimulation (direction of improvement may reflect reflexic vs. areflexic bowel)

Sexual function:

Hypothesis #17:  Epidural stimulation improves ability to achieve sexual arousal, erections, and ejaculation

Hypothesis #18:  Epidural stimulation leads to improved sexual satisfaction

Clinical testing and training

Clinical tests and assessments that would be appropriate for a study of the effects of epidural stimulation on bladder, bowel, and sexual function were considered. The expert consensus view was that a large team with multi-disciplinary input will be critical in designing and conducting a meaningful spinal stimulation study. In particular, specialists with clinical expertise in urology and gastrointestinal physiology should have a major role in the study design and execution. In addition, the team should include a full range of SCI specialty clinical experts (including SCI physiatrists, neurologists, neurosurgeons, radiologists, nurses, physical therapists, and rehabilitation psychologists), and critical input should be sought from neuroscientists, ethicists, regulatory specialists, statisticians, bioengineers, and people with SCI and their families or primary caregivers. Use of an experienced clinical study advisory board and an appropriate data and safety monitoring plan (DSMP) are highly recommended.

To achieve the study goals, it will be essential to obtain extensive baseline assessments of the participants prior to initiation of the intervention. These must be carefully considered to ensure inter-session and inter-rater reliability and to balance the need and desire to be thorough with the limitations of feasibility, time, and fatigability of the participants and the study team. SCI clinical assessments should include documentation of the neurological status, bladder, bowel and sexual functional status, and the extent and nature of the injury. To facilitate data collection and sharing, use of the SCI common data elements and the International SCI Data Sets is suggested. These have been developed by a large multidisciplinary team of clinical experts.

Existing capabilities to document the anatomical characteristics of the injury site and the functional capacity of the spinal cord before and during epidural stimulation were also considered. This included preliminary analysis of various imaging strategies suitable for addressing the known challenges of SCI imaging, including artifacts due to vertebral stabilization hardware and gating required to correct for respiratory and cardiac movement (Stroman et al., 2014). Magnetic resonance imaging (MRI) and diffusion tensor imaging (DTI) guidelines and parameters have been developed by SCI radiologists as part of the NINDS common data element (CDE) project, and these can be found at the NINDS CDE website. X-rays and CT could be used to precisely identify the location of electrodes and correlate with effectiveness. MRI compatible stimulation technology is evolving, and fMRI for investigational research might be feasible as part of the study design, pending appropriate regulatory approvals. The current commercially available spinal stimulators are still contraindicated for MRI, as the artifacts and risks of tissue heating are still unknown (Walsh et al., 2015). Thus, careful imaging of the spinal cord should be completed before implantation of a stimulator. The group also considered potential uses of functional spinal cord imaging to document activity patterns that might be altered by epidural stimulation. When available, MR compatible leads and stimulators should be strongly considered if they can enable post implant functional MR studies. Indeed, resting state functional MRI (fMRI) and DTI connectome tractography approaches show some promise for investigating mechanisms of stimulation device therapies in the brain (Sheline et al., 2010) yet functional imaging approaches are still exploratory for SCI indications and are limited by the size and other physical characteristics of the spinal cord. Some suggest the use of ultrasound as an effective tool for assessing bladder volumes and upper urinary tract changes. Also, a wide variety of highly informational electrodiagnostics tests can be performed prior to and during the intervention phase of the study to provide a broad picture of the functional status of spinal tracts and reflexes. 

Baseline measures and information that need to be taken before implantation include:

(1)  MRI imaging of the cord lesion, possibly fMRI of resting circuitry;

(2) Spinal cord conduction to evaluate completeness of lesion, e.g. somatosensory evoked potentials, motor evoked potentials, etc.; 

(3) Bladder tests to include urodynamics (filling cystometry and pressure flow studies including EMG  documenting bladder compliance; involuntary and voluntary bladder contractions;  and rhabdosphincter (or ‘external urethral sphincter’) contraction and relaxation; bladder capacity; volume voided; post-void residual volume; and sensation during filling);

(4) Use of a three-day bladder (catheterization and/or voiding) diary to document incontinence episodes, volume voided, frequency of voiding; obtain neurogenic bladder symptom score (Welk et al., 2014);

(5) Cardiovascular function in preparation for studying conditions, such as autonomic dysreflexia and postural hypotension; also  blood pressure and  autonomic dysreflexia monitoring during epidural stimulation in the lab;

(6) Medications being used that affect the bladder, bowel, and sexual function;

(7) Bladder management, including information about use of catheterization;

(8) Bowel diary, indicating fecal continence, dates and times of defecation and duration of bowel emptying activity and medications affecting bowel; obtain colonic transit time; obtain the bowel dysfunction score (Krogh et al., 2006);

(9) Sexual function questionnaire and erectile response to vibratory stimulation of penis or sacral segments.

For rapid preliminary assessment of changes in lower urinary tract function after stimulation, sphincter reflex measures should be considered with particular attention to bladder sphincter coordination during periods of urine storage and voiding. This will require urodynamics with a urethral pressure sensor, EMG and complete fluoroscopy.

The following for mapping bladder, bowel, and sexual function have been suggested: (1) Initial localization of potential sites can be determined by mapping stimulation sites for skeletal muscles that are known to be proximate to the bladder nuclei; (2) Recording of rapid onset and rapid offset responses that can be reproduced at short inter-stimulation intervals, which is preferable to examining the effect of stimulation on bladder filling and voiding responses (a very slow process) (Gustafson  et al., 2004; Horvath et al., 2010; Kennelly et al., 2011; Yoo et al., 2007; Yoo et al., 2011); (3) Recording of sites that promote urine storage and continence by recording phasic reflex bladder contractions in a partially filled bladder under isovolumetric conditions and then determining which sites suppress these contractions, or by recording sphincter EMG activity and determining which sites enhance this activity; (4) Mapping of spinal sites that promote voiding in participants with detrusor-sphincter-dyssynergy by examining spinal sites where stimulation suppresses guarding reflex activity; the longer term effects should be correlated with the rapid onset effects and placed on the map; (5) Mapping of smooth muscle response to single electrode stimulation with trains at about 20 Hz; (6) Modulation of thoracic and lumbosacral visceral reflexes by low level stimulation.  The visceral reflexes include sympathetic pathways arising in the thorocolumbar segments of the spinal cord.  Note that the modulation of sympathetic control of the bladder, urethra, distal bowel, anal canal and seminal emission could be an important component of epidural stimulation. In addition, suppression of autonomic dysreflexia may also involve modulation of thoracic and lumbar sympathetic pathways to blood vessels and heart; (7) Mapping of both animal and human subjects, which should be conducted in parallel (Kruse et al., 1991;Tai et al., 2000; Tai et al., 2004; Tai et al., 2006; Chang et al., 2006). 

Several commercial systems have ‘research modes’ with expanded stimulation capabilities that might be unlocked in appropriate IDE/IRB controlled trials. Investigators should work closely with manufacturers in planning and executing mapping studies in order to make effective use of these additional features. To help facilitate this interaction a conference was recently held at NIH (BRAIN Initiative Program for Industry Partnerships to Facilitate Early Access Neuromodulation and Recording Devices for Human Clinical Studies, June 3-4, 2015). Links to the agenda and videocasts of the presentations are available here.  

For all studies, it will be important to record the location and frequency of stimulation. It was recommended to map response to stimulation along the three coordinates of location, recording amplitude, and frequency. It was noted that it is important to identify the acute and long term effects of stimulation. There is a need to decide the optimal number of electrodes as well as the coverage of electrodes based on outcome.

Quality of life measures should be obtained to assess the impact of the intervention on the health and well-being of the participant over time. The current SCI-QOL instrument provides measures that assess such impact; this has become an essential consideration in determining the effectiveness of studies.

Careful consideration should be given to medications that affect bladder, bowel, or sexual function. It is recommended that patients with bladder botox are excluded from the study but those receiving anticholinergics and alpha blockers can be included because these medications will have little effect on general volitional voiding and may help with bladder capacity as long as the dose is steady. Botox takes up to one year to wear off so it is suggested that a criterion for inclusion is that participants should not have had treatment of botox to the bladder within a year.

Participant selection and length of follow-up

Participant selection should be specific to the hypotheses under study. When the hypothesis, for example, is that SCI individuals will experience improved bladder capacity after stimulation, researchers should include participants who are most likely to exhibit change (e.g. those with small bladder capacity measured during repeated baseline sessions). Participants should not be excluded based on their use of urinary catheters. Experts suggest that individuals should be at least two years out from their injury and stable without any other major health complications including cancer, diabetes, heart disease, or stroke. Participants should be willing and able to return for repeated follow-up studies as required to assess functional and health status.

Studies using epidural spinal stimulation in individuals with SCI have so far only included men. Women, however, make up 20 percent of all persons with SCI (2013, National Spinal Cord Injury Cord Injury Statistical Center). Due to a scarcity of research and the need to understand the differences, if any, in the response to epidural stimulation, these early feasibility studies should include both males and females. A 1998 study found differences between men and women in cause of injury, use of medications, attendants, transportation, and type of insurance. However they also found "more similarities than differences in the ways in which they manage life with SCI." (Shackelford, 1998). Efforts should be made to recruit under-represented minority populations to prospective studies.

Issues related to length of follow-up require further consideration. Stimulation can have an immediate impact on function. If an immediate effect is seen, follow-up will be needed to determine whether the effect persists and whether it grows stronger or weaker with time. Estimates of the time course of immediate effects will be needed to plan appropriate follow-up. It is possible that some effects of stimulation may only emerge after stimulation has been used for a long duration. This should be taken into consideration when determining the length of follow-up, as the study time should match the time course of the apparent effect.

There is also the possibility that after completing a study, subjects will choose to keep their stimulators implanted and continue to use them in daily life. Long-term follow-up for this population should be planned within the study design. For the long-term, it has been suggested that the researcher and his or her institution follow the participant every six months. Reasons for removing the device may include: the device is no longer serving the best interests of the participant, or the participant requests removal of the device.

The study protocol and informed consent document should address specific plans for long-term follow-up and maintenance of the spinal stimulation device. Long-term follow-up and maintenance should be offered by the study team or through appropriate referrals local to the study participant. This should involve consideration of long-term resource allocation. Initiation of an invasive device study includes a moral and financial obligation of the investigators to support the participant follow-up, regardless of the status of commercial entities or funding support.

Safety and anticipated adverse events

All human subject studies should be designed and performed under close supervision by an appropriate safety monitoring board comprised of individuals who are not directly involved in the research goals. With regard to general safety of participants in the lab and at home during the study, each participant will likely differ in terms of when they can be sent home with the stimulator. Researchers at The University of Louisville have developed a checklist that includes criteria about “readiness” of the participant to function independently outside the lab after training. When a participant is sent home, programs for the stimulator are limited so that the greatest degree of safety is assured for each participant. To protect the health and safety of the participant, it is essential that investigators work closely with device manufacturers, FDA reviewers, and the institutional IRB.

Given the long-term goal of providing bladder control without medication, it is recommended that interventions are conducted with and without medication for bladder control. However, patient safety should be thoroughly considered before taking patients off any medications.

Potential adverse events considered by the Consortium included: skin breakdown or infections at the site of the stimulator implant, autonomic dysreflexia, hypotension, hypertension, bladder storage issues, loss of sensation, bladder distension, spasticity in muscles of the legs, increased number of infections, constipation, and changes in voiding function.

It is recommended that blood pressure monitors be used during stimulation to check for autonomic dysreflexia at the lab and when the participant is at home. Researchers are also urged to monitor kidney function and perform renal ultrasound every six or seven months until measurements are stable. Renal function should be monitored thereafter on an annual basis.

The knowledge base provided by the Food and Drug Administration of potential adverse outcomes for commercial epidural stimulators for patients seeking relief from pain of trunk and limbs should also be considered.  Other sources describing possible adverse outcomes are Medtronic, Boston Scientific, St. Jude Medical (see highlight below), and Levy et al., 2011 (see reference list at end of document). 

Endpoints and evaluation of success

The initial studies are aimed at mapping to identify stimulation parameters that can produce a large change in bladder, bowel, or sexual function without any adverse effects. These initial studies will objectively determine the effectiveness of any changes that can be achieved and will provide some measure of the variability in change as well as variability in the choice of stimulation parameters. With improved bladder/bowel and sexual function as primary outcomes for the initial study, the Consortium explored briefly what might be enough objective change to be significant for the participant. It is recognized that the cost-benefit balance for changes in measures such as bladder capacity may be patient specific. The recently published FDA draft guidance on Patient Preference Information may be useful in helping clarify a ‘successful’ outcome for a specific patient. In any event outcomes for a definitive clinical trial will need to be defined after these initial mapping studies are complete.

Summary and future research considerations

There is a high level of enthusiasm for developing the potential of epidural stimulation as an intervention for people with SCI. There is also recognition that the mechanism(s) of action are not understood and that procedures for determining effective stimulation parameters in terms of the stimulation site, and stimulation intensity, rate and schedule are not well developed. In regard to planning future epidural stimulation studies, it was suggested that there be a small pilot study to determine the best set of stimulation parameters prior to initiation of a controlled trial. This would help address the pressing need for more information about the time course of the changes, the optimal sites for stimulation, the optimal patterns for stimulation and for ways to develop stimulation parameter space mapping efficiently for a specific individual. While it is probable that a series of basic principles can be identified in the selection of optimized stimulation parameters based on animal studies, there was general agreement that this eventually must be demonstrated in human volunteers. The details however, of how to best approach stimulation space mapping are challenging. The potential stimulation space is large and the time periods for many of the effects are long. Thus, a simple brute force approach is not realistic. The best chance of realizing an effective outcome in the spinal cord stimulation space is to encourage a concerted, multidisciplinary approach that utilizes scientific knowledge of the anatomy and physiology of the relevant systems. There is also a need to develop algorithms to systematically optimize the selection criteria for effective epidural stimulation.

Appendix A: Consortium Members

NIBIB Leadership and Staff