In-Clinic and At-Home Patient Experiences with Closed-Loop Spinal Cord Stimulation: 12-Month Outcomes from the Closed-Loop Australia Randomized Crossover Trial

Abstract

Introduction

Closed-loop spinal cord stimulation (CL-SCS) uses evoked compound action potential (ECAP) feedback to dynamically adjust stimulation amplitude, reducing under- and overstimulation during postural and activity changes. This manuscript reports 12-month patient-reported stimulation and pain outcomes from the Closed-Loop Australia study.

Methods

This multicenter, prospective, single-arm study used randomized in-clinic crossover testing to investigate the primary outcome of overstimulation in patients using CL-SCS versus open-loop (OL) SCS during protocol-specified movements. Secondary outcomes included patient-reported stimulation characteristics for patient-informed, at-home therapy and pain intensity.

Results

At 12 months, patients randomized to CL-SCS reported lower rates of overstimulation (median 0.4; IQR 0.0, 1.0) compared with OL-SCS (median 3; IQR 2.6, 4.0). Participants described the stimulation characteristics of their predominant at-home SCS therapy in terms of comfort and ability to perform activities. Eighty-five percent of participants (n = 39) reported being comfortable performing physical activity while their stimulation therapy was engaged, while 19.6% (n = 9) indicated needing to turn their therapy down or off to do certain activities. Pain intensity decreased from baseline by 60 mm and remained stable through 12 months, with 78% (n = 36) of participants achieving ≥ 50% pain relief.

Conclusions

These findings suggest that adaptable stimulation strategies are important in supporting comfort and sustained pain relief.

Trial Registration Number

ClinicalTrials.gov, NCT05177354.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1007/s40122-026-00863-6.

Key Summary Points

Why carry out this study?

Closed-loop spinal cord stimulation (CL-SCS) uses evoked compound action potential feedback to adjust stimulation amplitude in response to posture- and activity-related changes, potentially reducing inconsistent therapy delivery associated with open-loop SCS.

Patient experiences of under- or overstimulation during daily activities may contribute to discomfort, therapy adjustments, activity avoidance, or dissatisfaction with SCS therapy, highlighting the need to evaluate patient-centered stimulation experiences over longer follow-up.

This study asked whether ECAP-modulated CL-SCS would continue to reduce patient-reported overstimulation compared with OL-SCS during randomized in-clinic crossover testing at 12 months, while maintaining durable pain relief and favorable at-home stimulation experiences.

What was learned from the study?

At 12 months, patient-reported overstimulation during in-clinic testing was significantly lower with CL-SCS than with OL-SCS.

During at-home use, most study participants reported feeling comfortable performing daily activities with their therapy engaged, while a few reported needing to turn their SCS therapy down or off to perform certain activities.

Pain relief was durable through 12 months, with median VAS pain intensity reduced by 60 mm from baseline and 78.3% of participants achieving at least 50% pain relief.

Introduction

Novel closed-loop spinal cord stimulation (CL-SCS) systems represent an advancement over traditional, low-frequency, open-loop (OL-SCS) in relieving chronic pain [1, 2]. Prior studies have demonstrated that patients using conventional OL-SCS may experience inadequate or excessive stimulation during activities of daily living. Inconsistent SCS therapy dose may lead patients to compensate for these fluctuations manually, increasing the stimulation amplitude when they experience insufficient pain relief and decreasing amplitude when they encounter overstimulation [3]. In addition to the burden managing their SCS therapy with each activity, these adjustments may lead patients to experience sub-optimal therapy if stimulation amplitude is set below the therapeutic window [4]. A recent evaluation of a paresthesia-based CL-SCS system reported that the OL control arm frequently operated below the intended target amplitude, with the median amplitude falling short 49% of the time [5]. Patient experiences with under- and overstimulation can prompt fear-based avoidance of activities of daily living, dramatically decreasing quality of life, or eventual discontinuation of SCS therapy from the perception their therapy is ineffective [3, 6, 7].

To address the challenge of inconsistent dosing, CL-SCS systems have been developed to dynamically adjust stimulation amplitude even as distance between the spinal cord and therapy leads changes, maintaining more consistent stimulation delivered to patients. Closed-loop SCS devices use evoked compound action potentials (ECAPs) to detect the extent of nerve fiber activation by each SCS pulse, information that is used to determine the amplitude of the next SCS pulse delivered [4, 5, 8]. Additionally, ECAPs can be sensed at a fast rate (typically ≥ 50 times per second) and thus enable detection of rapid events that may not involve distinct changes in posture (e.g., cough, sneeze) to adjust stimulation. This ensures more consistent therapy delivery while reducing the risk of under- or overstimulation caused by changes in posture or activity [1, 2, 8].

Early 1- and 3-month outcomes from the Closed-Loop Australia study were previously reported, including detailed study methodology and patient-reported outcomes during in-clinic testing and early follow-up [9]. At 3 months, we found that 96% of participants had lower average in-clinic overstimulation intensity when using CL-SCS compared with OL-SCS. Given the known potential patient tolerance of stimulation or for therapy loss of efficacy, the objective of the present manuscript is to report long-term durability of the CL feature to prevent overstimulation. Here, we report the results of in-clinic testing 12 months post-implant using a CL system capable of multiplexing low and high frequency waveforms and compare overstimulation responses during in-clinic CL and OL crossover testing [10, 11]. Secondary outcomes include patient-reported stimulation experience and pain outcomes through 12 months of follow-up, with an emphasis on real-world stimulation characteristics during at-home therapy. We hypothesized that patient experiences of overstimulation during in-clinic testing would continue to favor ECAP-modulated CL-SCS for overstimulation reduction and patient-reported secondary outcomes would be significantly improved from baseline and remain robust across study visits.

Methods

Data and Study Design

The Closed-Loop Australia study is a multisite, prospective, single-arm feasibility study with randomized, blinded, in-clinic crossover testing comparing CL-SCS and OL-SCS during protocol-specified movements. It was registered on 30 November 2021 on ClinicalTrials.gov (NCT05177354). The Closed-Loop study was approved by Bellberry Human Research Ethics Committee on 09 November 2021 (Reference # 2021-08-959). The Closed-Loop study was designed and executed in accordance with the principles outlined in the Declaration of Helsinki and applicable regulatory requirements. All participants provided written informed consent prior to enrollment in the Closed-Loop study.

Eligible study participants were adults with chronic back, leg, and/or upper-limb pain and baseline visual analog scale (VAS) pain intensity ≥ 60 mm were enrolled (n = 94). Participants who achieved ≥ 50% pain relief during device trial were eligible to proceed Inceptiv™ implantation and long-term follow-up. The first patient was enrolled on 01 December 2021. Full details of the study protocol, randomization scheme, and blinding have been previously published [9]. Briefly, a visit-specific randomization listing using permuted blocks of 2 and 4 were generated to maintain balance between randomization sequence [9].

The study consisted of two distinct phases. First, the primary endpoint assessments of in-clinic overstimulation measured at 1, 3, and 12 months post-implant [9]. The second phase included long-term follow-up through 24 months to assess durability of pain relief, and impact of at-home treatment on quality of life and function. Data collected at 12 months included the final assessment of in-clinic overstimulation testing, self-reported pain intensity, and self-reported therapy experiences with predominant at-home therapy. The present analysis includes data collected through 12-month follow-up (n = 46), Fig. 1.

Fig. 1.

Study flowchart

Participants’ devices were programmed with therapy parameters that best managed their pain on the basis of in-clinic feedback. Additionally, in-clinic randomization used a participant’s optimized SCS therapy parameters with and without the closed-loop feature during test sequences. For those programmed with differential target multiplexed (DTM™), options included one low-rate (50 Hz) and one high-rate (≥ 200–900 Hz) program applied to two different targets. A detailed explanation of programing modalities has been previously published [9].

Measures

The primary outcome—overstimulation—was measured at the 3- and 12-month in-clinic visits. Study participants chose from a set of prescribed movements (e.g., back arch, cough, arm raise) three activities known to them to elicit a sensation of overstimulation. Change in ECAP amplitude during each activity was confirmed prior to a participant’s randomization assignment. A few participants reported only one or two prescribed activities elicited overstimulation; after confirmation by ECAP amplitude they were allowed to choose only relevant activities. Participants then repeated each chosen activity up to three times during the randomization test sequence with closed-loop on and with closed-loop off. Given heterogeneity of primary pain complaints among study participants, study coordinators did not impose uniformity in selected movements across patients [9]. After each movement, participants rated perceived overstimulation intensity on a 5-point scale ranging from none (0) to very strong (4). A mean overstimulation score was reported separately for movements performed with CL-SCS and with OL-SCS.

Secondary outcomes assessed patient-reported real-world stimulation characteristics and pain intensity using their programmed, at-home therapy. At the 12-month study visit participants self-reported real-world stimulation characteristics associated with their predominant at-home stimulation mode, including perception of stimulation, comfort with physical activity, fear of overstimulation with movement, and need to reduce or deactivate therapy to perform activities. Participants were asked if they felt any stimulation yes/no and asked to rate their at-home stimulation experiences on a 5-point scale dichotomized as “yes” (neutral/agree/strongly agree) or “no” (disagree/strongly disagree). Finally, patients reported pain intensity over the prior 72 hours using the visual analog scale (VAS) [12].

Analysis

The sample size calculation as well as the primary and secondary endpoint analysis and results at 1 and 3 months have been published previously [10]. Given that the primary endpoint of the study was measured at 1 month post-implant, all analyses of the primary and secondary outcomes reported herein are exploratory and therefore not explicitly powered. Within-participant comparisons of in-clinic overstimulation scores between CL-SCS and OL-SCS were conducted using Wilcoxon signed-rank tests. Descriptive statistics were used to assess secondary outcomes of pain intensity and self-reported, real-world stimulation characteristics, both overall and stratified by at-home SCS therapy. Overall change in average pain intensity from baseline through 3, 6, and 12 months are expressed as median, interquartile range.

Results

A total of n = 46 participants with implanted devices had data available at baseline and 12-month follow-up (Fig. 1). Reasons for attrition between the primary endpoint at 1 month (n = 55) and 12 months included physician (n = 5) or participant decision (n = 3) (e.g., development of new or confounding pain conditions, decision to pursue other therapy options), lost to follow-up (n = 2), and adverse event (n = 1). Participants in this study were more likely to be male, aged 50 or older, have both back and leg pain, and 80% (n = 37) predominantly used CL-SCS for at-home therapy compared to OL-SCS (n = 9). Among those using CL-SCS at home, 58.7% (n = 27) patients used CL DTM™ while 21% (n = 10) used low-dose (LD) or an alternative CL waveform (Table 1).

Table 1.

Summary of 12-month follow-up participant demographic and SCS therapy characteristics

Characteristic	N (%)
Age at enrollment, mean ± SD	59.1 ± 13.8
Female	20 (43.5)
Self-reported race
White	37 (80.4)
Asian	1 (2.2)
Black	0 (0.0)
Other	8 (17.4)
Primary location of pain at enrollment
Back only	10 (21.7)
Leg only	6 (13.0)
Back & leg	28 (60.9)
Upper limb	2 (4.3)
Responder rate (≥ 50% pain relief as reported on VAS)	36 (78.3)
SCS therapy modality used at home
Closed loop, any programminga	37 (80.4)
Closed-loop DTM waveform	27 (58.7)
Closed-loop LD waveform	8 (17.4)
Closed-loop other waveform	2 (4.4)
Open-loop, any programmingb	9 (19.6)

Demographic and programming characteristics of participants that completed the 12-month follow-up visit. Demographic characteristics are patient-reported at enrollment; responder rate and at-home SCS therapy modality are reported at 12-month follow-up. Data are presented as N (%) unless otherwise specified

DTM differential target multiplexed, LD low-dose, SCS spinal cord stimulation, VAS visual analog scale

^aWaveforms with closed-loop are characterized by one low-rate (50 Hz) and one high-rate (≥ 200–900 Hz) program applied to two different targets, pulse width 120, 150, or 200 µS; CL-LD: program frequency < 200 Hz excluding subwave parameters; CL-subwave: program frequency 100 Hz, pulse width 1000 µS; CL-other: can neither be characterized by parameters above nor high-density parameters (program frequency 1000 Hz, pulse width 90–220 µS)

^bOpen-loop programming waveforms are characterized as LD (n = 3): program frequency < 200 Hz; subwave (n = 4): program frequency 100 Hz, pulse width 1000 µS; and other (n = 2): can neither be characterized by parameters above nor high-density parameters (program frequency 1000 Hz, pulse width 90–220 µS)

In-Clinic Stimulation Sensation Testing

During in-clinic single-blind, randomized, crossover testing at 12 months, median overstimulation scores decreased significantly with CL-SCS (median 0.4; IQR 0, 1), with patients reporting no (0) or weak (1) overstimulation compared with OL-SCS (median 3.0; IQR 2.6, 4) where patients predominantly reported moderate to very strong overstimulation (3) during protocol-specified movements. Paired analysis confirmed both a statistically significantly lower rate of overstimulation with CL-SCS (p < 0.001) and a large effect size (r = 0.86) (Table 2) [13]. These findings are consistent with results observed at the 3-month follow-up visit.

Table 2.

Patient-reported stimulation sensation during randomized, crossover, in-clinic testing

SCS therapya	3 months (n = 50)b				12 months (n = 43)c
	Score median (IQR)	z score	p value	r	Score median (IQR)	z score	p value	r
CL-SCS	0.2 (0, 0.8)	− 6.04	< .001	0.87	0.4 (0,1)	− 5.5	< .001	0.86
OL-SCS	3 (2, 3.6)				3 (2.6, 4)

Patient-reported overstimulation scores during randomized, single-blind comparison of CL-SCS with OL-SCS are analyzed using Wilcoxon-signed rank test. Overstimulation scores decrease significantly with CL-SCS compared with OL-SCS at 3-month (z = − 6.04, p < 0.001) and 12-month (z = − 5.5, p < 0.001) follow-up visits. Overstimulation scores are presented as median (IQR). Overstimulation scale: 0, no stimulation sensation; 1: weak overstimulation sensation; 2: moderate overstimulation sensation; 3: strong overstimulation sensation; 4, very strong simulation sensation

CL closed-loop, IQR interquartile range, OL open-loop, SCS spinal cord stimulation

^aIn this crossover study design the same participants receive both treatments

^bNo ECAP reading available for n = 4 at 3-month follow-up

^cNo ECAP reading available for n = 3 at 12-month follow-up

Real-World Stimulation Characteristics

At 12-month follow-up, 84.8% (n = 39) patients self-reported being comfortable engaging in physical activity while using SCS therapy at-home. Few patients 19.6% (n = 9) reported needing to turn the therapy down or off to perform certain activities, while only 8.7% (n = 4) reported fear of overstimulation during everyday movements (Table 3). Patient characterization of stimulation was consistent from 3 to 12 months of follow-up.

Table 3.

Characteristics of spinal cord stimulation therapy reported at 3- and 12-month follow-up

Characterization of at-home SCS therapy	3-month follow-up (n = 54)		12-month follow-up (n = 46)
	Yes	No	Yes	No
“Did you feel the sensation of stimulation?”	43 (79.6)	11 (20.4)	33 (71.7)	13 (28.3)
“I notice that the therapy is on”	39 (72.2)	15 (27.8)	35 (76.1)	11 (23.9)
“I’m afraid that if I move in certain ways my therapy will shock me or become too intense”	14 (25.9)	40 (74.1)	4 (8.7)	42 (91.3)
“I’m comfortable being physically active with my stimulation therapy turned on”	43 (79.6)	11 (20.4)	39 (84.8)	7 (15.2)
“I have to turn down or turn off my stimulation therapy in order to do certain activities.”	14 (25.9)	40 (74.1)	9 (19.6)	37 (80.4)

Participants at the 3-month and 12-month follow-up visits characterized their SCS therapy relative to at-home use. Each question was answered with a 5-point Likert scale from “strongly disagree” to “strongly agree”. Responses are dichotomized into “yes” (strongly agree, agree, neutral) and “no” (strongly disagree, disagree) and summarized by visit, i.e., N (%)

SCS spinal cord stimulation

Pain Intensity

Median VAS-reported pain intensity decreased by 60 mm from baseline (pre-implant) through 12 months of follow-up (Table 4, Fig. 2). Reductions in pain intensity remain consistent throughout follow-up and represent a durable effect of pain relief over time. Seventy-eight percent (n = 36) of participants sustained ≥ 50% pain relief at 12 months.

Table 4.

Pain intensity reported on the visual analog scale from baseline through 12-month follow-up

Pain intensity	Baseline visit (n = 60) Median (IQR)	3-month follow-up (n = 54) Median (IQR)	6-month follow-up (n = 51) Median (IQR)	12-month follow-up (n = 46) Median (IQR)
VAS, mm	80.5 (73.75, 88.25)	16 (8, 30.5)	26 (8.5, 39.5)	20.5 (5.75, 32.75)

Pain intensity, as reported on the VAS (mm), is presented as median (IQR) and summarized by study visit. Patients marked a line best representing their level of low-back and/or leg-pain over the past 72 h on a paper scale from 0 mm (no pain) to 100 mm (worst pain imaginable)

IQR interquartile range, VAS visual analog scale

Fig. 2.

Median VAS-reported pain intensity from baseline (pre-implant) through 12 months of follow-up

Safety and Adverse Events

There were four device-related adverse events and three serious adverse device events reported pre-implant between enrollment through the end if the device trial (Supplemental Table 1). Post-implant through 12 months, 30 adverse device events were reported among 60 implanted participants and six serious adverse device events were reported. Device-related events are categorized as relating to the device, therapy, or implant procedure. Adverse event relatedness is determined by principal investigator. Between device trial and 12-month follow-up two serious adverse events occurred that led to early exit. One participant experienced a serious adverse event during the device trial (i.e., pre-implant) related to the trial device and procedure. The second participant experienced a serious adverse event post-implant prior to 3-month follow-up unrelated to device, therapy, or procedure. Both serious adverse events have been reported previously [9].

Discussion

This 12-month analysis suggests ECAP-modulated CL-SCS alleviates patient-reported overstimulation during blinded, randomized in-clinic testing, with more patients using CL-SCS, specifically the DTM™ waveform at-home, between study visits. We also find that most participants experience sustained pain relief through 12 months and descriptively report that their at-home SCS therapy did not limit physical activity related to activities of daily living. Collectively, these findings highlight the potential efficacy of CL-SCS for reducing overstimulation and illustrate CL-SCS integration for optimal pain management.

Closed-loop spinal cord stimulation appears to address a core limitation of open-loop therapy: stimulation that varies with posture and activity. The results of our in-clinic, crossover testing signal reduced overstimulation during movements designed to alter the distance between the leads and the spinal cord. These results are similar to earlier findings from the in-clinic ECHO-MAC crossover device trial where 97.6% reported reduced stimulation sensation while using CL-SCS [14]. The same study also showed lower ECAP variability with closed-loop, which supports the interpretation that the system is not just changing pain perception but keeping therapy delivery more stable and comfortable across changing body positions [14].

Descriptive data on at-home stimulation sensations were largely similar across patients using CL or OL-SCS, indicating that over- and understimulation during daily activities are common and may lead to avoidance of activities and manual adjustments to therapy [3]. In contrast to earlier patient experience surveys on patients using OL-SCS, we found that patients reported being comfortable engaging in physical activity while using their SCS therapy, with few reporting fear-based avoidance of movement related to overstimulation. Only nine patients reported adjusting or deactivating their therapy to perform certain activities, suggesting that stimulation delivery was compatible with activities of daily living and did not adversely limit patients.

In parallel with functional benefits, participants demonstrated sustained reductions in pain intensity through 12 months following implantation. Median pain scores decreased by 60 mm on the VAS from baseline and were sustained throughout follow-up. Over 78% participants implanted experienced 50% or greater pain relief at 12-months follow-up. These durable effects are consistent with other studies examining long-term pain relief, indicating CL SCS may be able to prevent sub-optimal therapeutic dose and dissatisfaction common to OL-SCS that leads to discontinuation [7, 15, 16].

Strengths and Limitations

Our study was designed as a short-term feasibility study of patient experience, and results support short-term efficacy of CL-SCS in both the durability of pain relief and patient satisfaction [9]. The in-clinic crossover design reduced interparticipant variability and increased our ability to detect treatment effects when directly comparing overstimulation sensations by SCS treatment within the same patients, thus reducing confounding. The extended follow-up posed logistical challenges for participants, including longer travel time to study sites and difficulty attending scheduled visits. Patients with a variety of primary pain complaints including low-back, leg, and upper-limb pain were also included to inform patient experience from a sample reflecting the broader chronic pain population. However, patient heterogeneity including differences in patient response to SCS treatment related to pain indications, follow-up duration, and logistical challenges led to higher than expected attrition. Sensitivity analyses were performed to assess the impact of attrition on outcomes; no significant differences due to censoring are found. Patient heterogeneity related to primary pain complaint also impacted the selection of movements designed to elicit overstimulation during in-clinic testing; as such, this may limit generalizability. Additionally, although our study did collect a blinding assessment, we did not formally assess the efficacy of blinding. These analyses represent secondary endpoints over long-term follow-up. As such, they are exploratory and should be interpreted with caution. Our feasibility design prioritized patient flexibility and individual treatment needs when determining at-home treatment. Multiple waveforms were available to be programmed for both OL-SCS and CL-SCS for at-home use; system utilization during at-home use was not pre-specified and therefore was not reported. Finally, our at-home stimulation characterization measure was self-reported and could have introduced recall bias. The methodological limitations of this study are being considered in the context of planned future studies.

Conclusion

This 12-month analysis demonstrates that during in-clinic testing, ECAP-modulated CL-SCS reduces overstimulation events. We also found durable long-term pain relief, associated with favorable patient reported stimulation experiences at-home SCS use. While descriptive, these outcomes illustrate the real-world, functional benefits of CL-SCS, which may not be fully reflected by pain intensity measures alone. Integration of patient-centered outcomes with adaptable stimulation strategies is important in supporting comfort and sustained analgesia long-term.

Supplementary Information

Below is the link to the electronic supplementary material.

Author Contributions

Vahid Mohabbati, Richard Sullivan, James Yu, Peter Georgius, Charles D. Brooker: writing-review & editing, investigation. Malgorzata Siorek: writing-review & editing, validation. Bonnie Ditterline: writing-review & editing, writing-original draft conceptualization, project administration, methodology. Jessica Friedman: writing-original draft, writing-review & editing, conceptualization. Marc A. Russo: Writing-Review & Editing, Investigation. All authors participated in the preparation, critical review, and approval of the manuscript, and the decision to submit for publication.

Funding

This study and the journal’s Rapid Service Fee was sponsored by Medtronic Inc, Fridley, MN, USA.

Data Availability

Primary and secondary endpoints are posted on ClinicalTrials.gov (https://clinicaltrials.gov/study/NCT05177354). The datasets analyzed in this manuscript are not publicly available to protect intellectual property.

Declarations

Conflict of Interest

Vahid Mohabbati and Marc A. Russo disclose receiving consulting fees and honoraria from the study sponsor relating to activities outside of this study. Vahid Mohabbati, Marc A. Russo, Charles D. Brooker, James Yu, Richard Sullivan, and Peter Georgius are investigators in the study and as such receive payments at fair-market value from the sponsor for study related activities. Marc A. Russo reports leadership roles in professional neuromodulation societies and equity interests in medical device companies. Richard Sullivan reports receiving honoraria for teaching and educational activities outside the submitted work. Peter Georgius reports consulting and advisory roles, travel support, and intellectual property interests outside the submitted work. James Yu reports a leadership role with the Neuromodulation Society of Australia and New Zealand. Bonnie Ditterline, Malgorzata Siorek and Jessica Friedman are employees of Medtronic.

Ethical Approval

The Closed-Loop study was approved by Bellberry Human Research Ethics Committee on 09 November 2021 (Reference # 2021-08-959). The Closed-Loop study was designed and executed in accordance with the principles outlined in the Declaration of Helsinki and applicable regulatory requirements. All participants provided written informed consent prior to enrollment in the Closed-Loop study.