(225 days)
Not Found
No
The description mentions a microprocessor running embedded software and an ASIC, but there is no mention of AI or ML algorithms being used for therapy control, data analysis, or any other function. The mobile app allows user control, not automated decision-making based on AI/ML.
Yes
The device is indicated as an aid for reducing the symptoms of fibromyalgia, which indicates a therapeutic purpose.
No
The device is a transcutaneous electrical nerve stimulation (TENS) device intended to reduce symptoms of fibromyalgia, which is a therapeutic function, not a diagnostic one. It delivers electrical stimulation to the user to aid in symptom reduction, rather than identifying or monitoring a medical condition.
No
The device description clearly states that the Quell-FM is a wearable device that includes hardware components such as a microprocessor, ASIC, battery, band, and electrodes, in addition to the mobile app software.
Based on the provided information, the Ouell-FM device is not an In Vitro Diagnostic (IVD).
Here's why:
- Intended Use: The intended use is to reduce the symptoms of fibromyalgia through transcutaneous electrical nerve stimulation. This is a therapeutic function, not a diagnostic one.
- Device Description: The device delivers electrical stimulation to the body. It does not analyze samples of human origin (like blood, urine, or tissue) to provide diagnostic information.
- Mechanism of Action: TENS works by stimulating nerves to potentially block pain signals or release endorphins. This is a physical intervention, not an in vitro test.
IVD devices are specifically designed to examine specimens derived from the human body to provide information for the diagnosis, monitoring, or treatment of diseases or conditions. The Ouell-FM does not perform this function.
N/A
Intended Use / Indications for Use
The Ouell-FM is a transcutaneous electrical nerve stimulation (TENS) device indicated as an aid for reducing the symptoms of fibromyalgia in adults with high pain sensitivity. The Quell-FM may be used during sleep. The Quell-FM is labeled for use only with compatible NeuroMetrix electrodes.
Product codes (comma separated list FDA assigned to the subject device)
OSO
Device Description
Quell-FM is a wearable, transcutaneous electrical nerve stimulator designed to stimulate sensory nerves in the upper-calf region. The device utilizes a microprocessor running embedded software and a custom high-voltage Application Specific Integrated Circuit (ASIC) to generate current regulated stimulating pulses with specific characteristics including pulse shape, amplitude, duration, pattern, and frequency. The device utilizes Bluetooth® low energy (BLE) to communicate with a mobile device that allows the user to start and stop therapy, control stimulation intensity, and modify certain operating characteristics. The device is powered by an embedded rechargeable lithium-ion polymer battery that is charged through a USB cable connected to an AC adapter. An image of the device and its placement on the upper calf is provided in Figure 1.
The primary components of the device include the Quell-FM device, Band, Electrodes, and Quell-FM mobile app.
-
A. Quell-FM Device
The Quell-FM device delivers electrical stimulation to the user through a disposable electrode placed on the user's body. The Quell-FM is labeled for use only with compatible NeuroMetrix electrodes (previously cleared in K140586), to which it connects through insulated female medical snap connectors embedded within its housing; no lead-wires are used. -
B. Band
A flexible band secures the Quell-FM device and the electrode to the user's leg using a hook and loop material. -
C. Electrodes
The Quell-FM device is labeled for use only with compatible NeuroMetrix electrodes (i.e., electrodes cleared under K140586). This use specification, in part, ensures the safe use of the device during sleep because NeuroMetrix electrodes have a known surface area that allows the device to quantitively determine relative skin contact area. Stimulation will be automatically stopped if device detects a decrease in skin-contact area which may lead to unsafe current density to be delivered as would occur during unattended use such as sleeping. -
D. Quell-FM Mobile App
Quell-FM is used with a mobile app, running on an iOS or Android mobile device, to which it communicates via Bluetooth. Using the mobile app, the user can start and stop the therapy, control stimulation intensity, and modify certain operating characteristics.
Mentions image processing
Not Found
Mentions AI, DNN, or ML
Not Found
Input Imaging Modality
Not Found
Anatomical Site
upper-calf region
Indicated Patient Age Range
adults, age 21 or older
Intended User / Care Setting
user, home healthcare environment
Description of the training set, sample size, data source, and annotation protocol
Not Found
Description of the test set, sample size, data source, and annotation protocol
A 119-subject clinical study was conducted. Subjects were randomized to an active or sham device with equal allocation. Active and sham devices were physically identical: only differing in whether they were loaded with standard software or modified software that implemented a sham stimulation protocol. Subjects were categorized into lower ( median) pain sensitivity subgroups based on baseline Quantitative Sensory Testing (QST) data using principal component analysis. The study evaluated the effectiveness and safety of the Quell-FM by comparing 3-months of at-home treatment with either the standard (active) or modified (sham) Quell-FM in individuals with fibromyalgia.
Summary of Performance Studies (study type, sample size, AUC, MRMC, standalone performance, key results)
Study Type: Double-blind, randomized, sham-controlled clinical trial.
Sample Size: 119 subjects (62 active, 57 sham). 100 subjects completed 3-months of treatment.
Key Results:
- Primary Effectiveness Outcome (PGIC): The difference in LS mean PGIC scores between active (3.54, SE 0.25) and sham (3.14, SE 0.26) treatment at 3-months was not significant in the ITT population (mean difference 0.40, 95% CI [-0.33, 1.13], p=0.279). In the higher pain sensitivity subgroup, PGIC was significantly greater for active treatment compared to sham treatment (mean difference 1.25, 95% CI [0.25, 2.24], p=0.015).
- Secondary Effectiveness Endpoints (ITT population):
- FIQR Total Score: Significant improvement with active treatment (mean difference -8.58, 95% CI [-15.44, -1.72], p=0.015).
- BPI Interference: Significant improvement with active treatment (mean difference -0.86, 95% CI [-1.64, -0.08], p=0.031).
- PDQ: Significant improvement with active treatment (mean difference -2.21, 95% CI [-3.90, -0.15], p=0.027).
- PDI: Significant improvement with active treatment (mean difference -4.90, 95% CI [-9.65, -0.14], p=0.044).
- Secondary Effectiveness Endpoints (High Pain Sensitivity Subgroup):
- FIQR Total Score: Significant improvement with active treatment (mean difference -11.31, 95% CI [-21.55, -1.07], p=0.031).
- FIQR Pain Item: Significant improvement with active treatment (mean difference -1.57, 95% CI [-2.59, -0.54], p=0.003).
- BPI Severity: Significant improvement with active treatment (mean difference -0.99, 95% CI [-1.90, -0.07], p=0.035).
- PDQ: Significant improvement with active treatment (mean difference -3.35, 95% CI [-6.13, -0.58], p=0.018).
- Responder Analyses:
- PGIC Responder (score ≥ 5): Active treatment: 42.4% (ITT), 57.8% (Higher Pain Sensitivity subgroup). The difference in responder rate between active and sham was 27.7% (p=0.025) in the higher pain sensitivity subgroup.
- FIQR Responder (≥ 15% reduction): Active treatment: 56.9% (ITT), 57.5% (Higher Pain Sensitivity subgroup). Differences between active and sham were 22.9% (p=0.014) in ITT and 29.5% (p=0.019) in the higher pain sensitivity subgroup.
- Pain Intensity Responder (FIQR pain item):
- ≥ 30% reduction: Active treatment: 45.8% (ITT), 59.5% (Higher Pain Sensitivity subgroup). Difference active vs. sham: 16.5% (ITT, p=0.074), 41.9% (Higher Pain Sensitivity subgroup, p
N/A
0
DE NOVO CLASSIFICATION REQUEST FOR QUELL-FIBROMYALGIA (QUELL-FM)
REGULATORY INFORMATION
FDA identifies this generic type of device as:
Transcutaneous electrical nerve stimulator to treat fibromvalgia symptoms. A transcutaneous electrical nerve stimulator to treat fibromyalgia symptoms is a prescription device that transcutaneously stimulates a patient's sensory nerves through electrodes placed on the skin.
NEW REGULATION NUMBER: 21 CFR 882.5888
CLASSIFICATION: Class II
PRODUCT CODE: OSO
BACKGROUND
DEVICE NAME: Quell-Fibromyalgia (Quell-FM)
SUBMISSION NUMBER: DEN210046
DATE OF DE NOVO: October 6, 2021
NeuroMetrix Inc. CONTACT: 4b Gill Street Woburn, MA 01801
INDICATIONS FOR USE
The Ouell-FM is a transcutaneous electrical nerve stimulation (TENS) device indicated as an aid for reducing the symptoms of fibromyalgia in adults with high pain sensitivity. The Quell-FM may be used during sleep. The Quell-FM is labeled for use only with compatible NeuroMetrix electrodes.
LIMITATIONS
The sale, distribution, and use of the Quell-FM are restricted to prescription use in accordance with 21 CFR 801.109.
Many participants in the clinical study were also taking medication for fibromyalgia and it was difficult to assess the effects of the device compared to medication.
1
The device is contraindicated for use by patients who have a cardiac pacemaker, implanted defibrillator, other implanted electronic device, or implanted metal near the device, because this may cause electric shock, burns, electrical interference, or death.
The stimulation electrodes should not be placed across or through the head, directly on the eyes, covering the mouth, on the front of the neck, on the chest or upper back, or crossing the heart.
The device cannot be used while driving, operating machinery, or during any activity in which electrical stimulation can put the patient at risk of injury.
PLEASE REFER TO THE LABELING FOR A MORE COMPLETE LIST OF CONTRAINDICATIONS, WARNINGS, AND PRECAUTIONS.
DEVICE DESCRIPTION
Ouell-FM is a wearable, transcutaneous electrical nerve stimulator designed to stimulate sensory nerves in the upper-calf region. The device utilizes a microprocessor running embedded software and a custom high-voltage Application Specific Integrated Circuit (ASIC) to generate current regulated stimulating pulses with specific characteristics including pulse shape, amplitude,
duration, pattern, and frequency. The device utilizes Bluetooth® low energy (BLE) to communicate with a mobile device that allows the user to start and stop therapy, control stimulation intensity, and modify certain operating characteristics. The device is powered by an embedded rechargeable lithium-ion polymer battery that is charged through a USB cable connected to an AC adapter. An image of the device and its placement on the upper calf is provided in Figure 1.
2
Image /page/2/Picture/0 description: The image shows three different images labeled A, B, and C. Image A shows a black device with a blue outline next to a black box with a blue outline. Image B shows a black strap with a blue circle in the middle and a hand placing a white circle on the blue circle. Image C shows a person putting the black strap with the blue circle around their knee.
Figure 1. Ouell-FM device and Band (A). Placement of device in Band and attachment of electrode (B). Placement of Quell-FM on upper calf (C).
The primary components of the device include the Quell-FM device, Band, Electrodes, and Quell-FM mobile app.
-
A. Quell-FM Device
The Quell-FM device delivers electrical stimulation to the user through a disposable electrode placed on the user's body. The Quell-FM is labeled for use only with compatible NeuroMetrix electrodes (previously cleared in K140586), to which it connects through insulated female medical snap connectors embedded within its housing; no lead-wires are used. -
B. Band
A flexible band secures the Quell-FM device and the electrode to the user's leg using a hook and loop material. -
C. Electrodes
The Quell-FM device is labeled for use only with compatible NeuroMetrix electrodes (i.e., electrodes cleared under K140586). This use specification, in part, ensures the safe use of the device during sleep because NeuroMetrix electrodes have a known surface area that allows the device to quantitively determine relative skin contact area. Stimulation will be automatically stopped if device detects a decrease in skin-contact area which may lead to unsafe current density to be delivered as would occur during unattended use such as sleeping. -
D. Quell-FM Mobile App
Quell-FM is used with a mobile app, running on an iOS or Android mobile device, to which it communicates via Bluetooth. Using the mobile app, the user can start and stop the therapy, control stimulation intensity, and modify certain operating characteristics.
3
The device has a single output mode consisting of continuous stimulation at a randomly varying instantaneous frequency centered at 40 Hz (range 30 - 50 Hz), 80 Hz (default, range 60 - 100 Hz) or 160 Hz (range 120 - 200 Hz). The technical specifications are listed in Table 1.
| Specification | Value |
|---|---|
| Configuration | 1 stimulator (2-lead) |
| Waveform (e.g., pulsed monophasic, biphasic) | Biphasic with alternating leading phase, asymmetrical |
| Shape (e.g., rectangular, spike, rectified sinusoidal) | Rectangular |
| Regulated current or voltage | current |
| Maximum Output Voltage (+/- 5%) | 50 V @ 500 Ω |
| 118 V @ 2000 Ω | |
| 118 V @ 10000 Ω | |
| Maximum Output Current (+/- 10%) | 100 mA @ 500 Ω |
| 61 mA @ 2000 Ω | |
| 12 mA @ 10000 Ω | |
| Duration of primary (depolarizing) phase | 100 µs |
| Pulse Duration | 280 µs (does not include 100 µs inter-phase delay) |
| Pulse Frequency | 60-100 Hz (default, randomly varying) |
| 30-50 Hz (randomly varying) or | |
| 120-200 Hz (randomly varying) | |
| Net Charge per pulse (If zero, state method of achieving zero net charge.) | Normally 8 µC @ 500 Ω per pulse; |
| Normally 0 µC per sequential pair of pulses; | |
| zero net current | |
| Maximum Phase Charge | 18 µC @ 500 Ω |
| 18 µC @ 1000 Ω | |
| Maximum Current Density (r.m.s), Calculated for minimum electrode area of 28 cm² | 0.54 mA/cm² @ 500 Ω @80 Hz |
| 0.76 mA/cm² @ 500 Ω @160 Hz | |
| Maximum Average Current | 2.2 mA @ 500 Ω @80 Hz |
| 4.5 mA @ 500 Ω @160 Hz | |
| Maximum Average Power Density, Calculated for minimum electrode area of 28 cm² | 4.0 mW/cm² @ 500 Ω @80 Hz |
| 8.0 mW/cm² @ 500 Ω @160 Hz |
| Table 1. Quell-FM Output Parameters | |||
|---|---|---|---|
| -- | -- | -- | ------------------------------------- |
PROCEDURE DESCRIPTION
Quell-FM delivers therapy automatically as repeating 1-hour sessions with a 1-hour gap between sessions (or 30 minutes sessions and 30 minutes gap), as long as the device is on the body, including overnight. Quell-FM provides continuous stimulation during each session. The stimulation intensity is initially set to a strong but comfortable level through the calibration procedure and can subsequently be manually adjusted by users. The intensity increases slowly during a therapy session (starting after 10 minutes) to compensate for nerve desensitization to
4
electrical stimulation (habituation). By default, the intensity increases 0.3% per minute in a stepwise fashion (i.e., increase by 0.3% at one-minute intervals); this is equivalent to a 16% increase over a 60-minute therapy session.
Quell-FM can be used during wakefulness or sleep. Stimulation will stop if the device detects that stimulation cannot be correctly and safely delivered. i.e. "trip condition" (Table 2). The user may halt stimulation at any time using the mobile application or by double tapping the device case. Alternatively, stimulation will stop after the therapy session timer (typically 60minutes) has elapsed. The device has an optional auto-restart timer. If it is enabled then a therapy session will automatically start 1 hour (30-minute option is also available) following the end of the previous session, provided that the user did not manually halt stimulation during the previous therapy session, or a trip condition did not occur.
| Trip Condition | Description of Triggering
Criteria | Purpose |
|---------------------|---------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| No Load | Device not connected to patient | Prevent stimulation with maximum output
voltage due to open circuit. |
| Insufficient Charge | Charge delivered during
stimulation below target | Prevent stimulation that may be sub-
therapeutic. |
| Over Load | Charge delivered during
stimulation above target | Prevent stimulation that may be exceed
specification limits. |
| Electrode Peel | Electrode dislodging from skin | Prevent high current density due to small
electrode area resulting from unrecognized
electrode peeling such as during sleeping. |
| Low Battery | Insufficient battery charge to
start ( 4 on an 11-point numerical rating scale (NRS).
Exclusion Criteria
- . Diagnosis of cancer or any other malignant disease
- . Acute osteomyelitis or acute bone disease
- . Present or past Diagnostic and Statistical Manual of Mental Disorders (DSM-V) diagnosis of schizophrenia, delusional disorder, psychotic disorder, or dissociative disorder judged to interfere with study participation
- · Pregnancy
- . Any clinically unstable systemic illness judged to interfere with treatment
- A pain condition requiring urgent surgery .
- . An active substance use disorder, such as cocaine or IV heroin use (positive on the Mini International Neuropsychiatric Interview: M.I.N.I. v.5.0), that would interfere with study participation
- Have an implanted cardiac pacemaker, defibrillator, or other implanted electronic . device
8
c. Randomization
Subjects were randomized to an active or sham device with equal allocation. Active and sham devices were physically identical: only differing in whether thev were loaded with standard software or modified software that implemented a sham stimulation protocol. The study coordinators and investigators could not determine whether a device was an active or sham device based on markings or physical characteristics and did not discuss the stimulation experience with subjects. Subjects were told that two types of TENS were being evaluated, a "low intensity" device and a "high intensity" device. Blinding effectiveness for subjects and study coordinators was assessed at the end of the study.
d. Quantitative Sensory Testing
The Quantitative Sensory Testing (QST) procedures used to identify subjects diagnosed with fibromyalgia that had lower vs. higher pain sensitivity included mechanical and cold stimuli. Responses to punctate mechanical stimuli were measured using a standard set of weighted probes. Singular taps were performed on the metacarpophalangeal joint of the middle finger. Mechanical temporal summation was defined as the increase in pain from the first to the tenth stimulus. A pressure algometer was used to measure pain pressure thresholds (PPT) at . the trapezius muscle and thumb joint. Cuff algometry at the calf was used to assess responses to sustained mechanical pressure. Responses to noxious cold were evaluated using a repeated cold pressor task, which involved immersion of the right hand in a circulating water bath maintained at 4°C. Conditioned pain modulation was measured by assessing PPT at the trapezius during the water bath immersions.
e. Intervention
The standard Quell-FM device (active) provided 60-minutes of continuous stimulation during each 1-hour therapy session. The modified Quell-FM device (sham) provided three 2-minute periods of stimulation during each session (at 0. 28, and 58 minutes) for a total of 6-minutes of stimulation. The device placement on the upper calf and usage instructions were identical for the two devices. Subjects were instructed to maintain a strong but comfortable stimulation intensity and to use their device for at least two I-hour therapy sessions each day over the course of the study.
All subjects were asked to continue their pre-study analgesic medications with changes tracked through a weekly interview.
f. Study Endpoints:
Safety:
Adverse events were assessed in weekly phone calls with subjects.
Effectiveness:
The clinical study included one pre-specified primary effectiveness measure and seven prespecified secondary effectiveness measures as shown in Table 3.
Table 3. Pre-specified primary and secondary effectiveness measures
| Effectiveness Measures | Description |
|---|---|
| Primary | |
| Patient's Global Impression | |
| of Change (PGIC) | Overall belief about the effectiveness of treatment. This study included two |
| PGIC measures per Hurst and Bolton, a 7-point categorical verbal scale | |
| (PGIC-VRS) and an 11-point numerical rating scale (PGIC-NRS). The | |
| VRS ranged from (1) "no change or condition has gotten worse" to (7) "a | |
| great deal better and a considerable improvement that has made all the | |
| difference." The use of PGIC-VRS was prespecified in the protocol and | |
| included in the statistical analyses. | |
| Secondary | |
| Fibromyalgia Impact | |
| Questionnaire Revised | |
| (FIQR) | Fibromyalgia specific instrument for assessment of disease impact and |
| health-related quality-of-life (QoL). It is comprised of 21 individual items | |
| each rated on an 11-point NRS. The FIQR Total Score ranges from 0 to | |
| 100, with higher scores indicating more fibromyalgia related impairment. | |
| FIQR is composed of three subscales: function, overall impact and | |
| symptoms. The symptom subscale includes a pain intensity assessment | |
| (FIQR pain item) that may be analyzed separately and was used in the | |
| responder analyses. | |
| Brief Pain Inventory Short | |
| Form (BPI) | Assessment of pain severity and interference rated on an 11-point NRS. |
| BPI Severity is the average of 4 pain intensity items. Pain Interference is | |
| the average of 7 function items. BPI Severity and BPI Interference were | |
| analyzed as distinct effectiveness measures. | |
| painDETECT Questionnaire | |
| (PDQ) | Assessment of the presence and severity of neuropathic pain. The |
| painDETECT questionnaire is composed of 7 pain-quality items and 2 | |
| items for pain-course and pain-radiation. Recent studies have shown that | |
| the 7-item painDETECT questionnaire has better discrimination for | |
| neuropathic pain compared to the full 9-item instrument and was therefore | |
| used in this study. The 7-item score ranges from 0 to 35, with higher scores | |
| indicating greater neuropathic pain. | |
| Pain Disability Index (PDI) | Assessment of pain related disability. Score ranges from 0 to 70, with |
| higher scores indicating greater pain related disability. | |
| Hospital Anxiety and | |
| Depression Scale (HADS) | Assesses the severity of anxious and depressive symptoms. Score ranges |
| from 0 to 42, with higher scores indicating greater severity of anxiety and | |
| depression. HADS is composed of two subscales: anxiety and depression. | |
| Pain Catastrophizing Scale | |
| (PCS) | Assessment of pain rumination, magnification, and helplessness. Score |
| ranges from 0 to 52, with higher scores indicating greater catastrophic | |
| thinking. |
9
g. Statistical Analysis
The primary analysis of treatment effects was conducted in the intention-to-treat (ITT) population, which included all 119 randomized subjects.
10
One pre-specified subgroup analysis was carried out in the ITT population. The study protocol predicted that subjects with higher pain sensitivity based on OST measures would demonstrate the greatest treatment effects. Subjects were classified as having lower or higher pain sensitivity by their baseline OST data. Principal component analysis was applied to the correlation matrix of the baseline OST data to identify the prominent pain patterns in the study population. The first principal component accounted for 40% of the total variance. The component had negative loadings for mechanical pain thresholds, positive loadings for pain responses to punctate stimuli and cold stimuli, and a positive loading for temporal summation, and could therefore be interpreted as a composite index of pain sensitivity. Subjects were classified as lower ( median) pain sensitivity using this principal component. This yielded a lower pain sensitivity subgroup with 59 subjects and a higher pain sensitivity subgroup with 60 subjects.
The 3-month least-squares (LS) mean PGIC scores were compared between the active and sham treatment groups, controlling for baseline pain severity (BPI average pain item), tenderness (dichotomized FIOR tenderness item) and body mass index (BMI) with an ANCOVA model. Missing scores were filled in using single imputation. Significance was assessed by the two-sample t-test with a Type I error rate of 0.05 (two-sided). The subgroup analysis was conducted to test for heterogeneity in the treatment response based on pain. sensitivity. Between and within subgroup treatment effects were assessed with the primary ANCOVA model that included a treatment by subgroup interaction.
The baseline to 3-month LS mean change scores for the secondary effectiveness measures were compared between the active and sham treatment groups. controlling for baseline value, pain severity, tenderness and BMI with an ANCOVA model. Missing data were filled in using multiple imputation. Significance was assessed by the two-sample t-test with a Type I error rate of 0.05 (two-sided).
Responder analyses for PGIC, FIOR Total Score and pain intensity (FIOR pain item) were conducted to inform the clinical meaningfulness of treatment effects. Responder rates were compared between treatment groups using logistic regression, controlling for baseline pain severity, tenderness and BMI. P-values were not assessed against a significance threshold because the study was not specifically powered for responder analyses.
h. Results
Of the 170 individuals screened for the study, 119 met the inclusion/exclusion criteria and were randomized to an active (62) or a sham (57) device for 3-months. Among these subjects, 19 (10 active, 9 sham) withdrew: 16 (7 active, 9 sham) were lost to follow-up and 3 (3 active. 0 sham) withdrew but completed the 3-month assessments. The remaining 100 subjects completed 3-months of treatment, however 4 (1 active, 3 sham) did not return the 3month assessments via mail following implementation of COVID-19 restrictions on inperson clinic visits.
Analgesic Use
At the baseline, half of the subjects (010 %) were taking over-the-counter analgesics 11/6 were prescribed neuroleptics. 100 were prescribed an antidepressant and PM 1% were taking an opioid, including tramadol. At the baseline, 11 % and 1 1 % of the participants in the sham and active treatment groups use pain medication. There was no discernable change
11
in analgesic use over the course of the study and there were no significant differences between the treatment groups for any week.
Device Use
Table 4 displays key characteristics of the distributions for Therapy Hours and Wear Time (Hours) for subjects randomized to receive the active device or sham device. One therapy hour (also called therapy session) equals 1 hour of electrical stimulation. Wear Time represents the total amount of time the device is on the patient's skin and is approximated as (2x Therapy Hours - 1 hour) rounded down to the nearest hour.
| Sham Arm (n=59) | Active Arm (n=60) | All Subjects (N=119) | |
|---|---|---|---|
| Therapy Hours* | |||
| Median | 3.9 | 3.8 | 3.8 |
| 75th Percentile | 5.4 | 5.8 | 5.6 |
| 95th Percentile | 8.4 | 8.4 | 8.4 |
| Wear Time (Hours)* | |||
| Median | 6 | 6 | 6 |
| 75th Percentile | 9 | 10 | 10 |
| 95th Percentile | 15 | 15 | 15 |
Table 4. Distribution of daily therapy hours for subjects randomized to the active device or sham device
*One therapy hour (also called therapy session) represents 1-hour of electrical stimulation.
- Wear time represents the total amount of time the device is on the patient's skin and is approximated as (2 x Therapy Hours - 1 hour) rounded down to the nearest hour.
Blinding Assessment
Blinding was assessed as each subject completed the study by asking the coordinators and subjects to identify whether a low intensity or high intensity Quell-FM device was used. The coordinator identified the correct treatment in 54.7% (95% CI [45.2, 64.2]) of the 103 subjects that completed the study (n=100) or withdrew but provided the 3-month assessment (n=3). The treatment was correctly identified 63.8% (95% CI [51.4. 76.21) of the time for the active device and 43.8% (95% CI [29.7%, 57.8%]) of the time for the sham device.
Of the 99 subjects that completed the 3-month assessment, 86 answered the blinding question which was included in the satisfaction questionnaire. Among the 13 that did not answer this question, 3 did not complete any part of the satisfaction questionnaire. In the subjects that answered the blinding question. 50.0% (95% CI [39.4. 60.6]) identified the correct treatment. Subjects in the active group correctly identified their treatment 17.4% (95% CI [6.4. 28.3]) of the time and subjects in the sham group correctly identified their treatment 87.5% (95% CI
12
[77.3, 97.7]) of the time. Among all subjects, 84.9% (95% CI [77.3, 92.5]) believed they received a low intensity device.
Safety Endpoint
A total of 12 (5 active, 7 sham) adverse events were reported. They included rash at the site of the device, numbness and tingling, and muscle cramping. Six (3 active, 3 sham) were determined to be related to TENS use, 3 (1 active, 2 sham) were deemed possibly related to TENS use, and 3 (1 active. 2 sham) were judged to be unrelated to TENS use by the principal investigator. The 9 events that were definitely or possibly related to TENS use were minor and self-limited. The most common occurrence was a rash under the Quell- FM electrodes. Ten subjects averaged more than 8 therapy hours (i.e., about 15 wear hours) per day (Table 4). Among these ten subjects, one experienced skin irritation that resolved after 1-week of not wearing their device. These ten subjects did not report any other adverse events. Skin irritation is a known minor risk of TENS use that generally resolves with conservative measures.
Primary Effectiveness Outcome
The difference in the LS mean PGIC scores between active (3.54, SE 0.25) and sham (3.14, SE 0.26) treatment at 3-months was not significant in the ITT population (mean difference 0.40, 95% C1 [-0.33, 1.13], p=0.279) (Figure 2), In the pre-specified subgroup analysis, the interaction between treatment and baseline pain sensitivity was significant (p=0.020), which indicated that baseline pain sensitivity moderated the relationship between treatment and PGIC at 3-months. In the higher pain sensitivity subgroup. PGIC was significantly greater for active treatment compared to sham treatment (mean difference 1.25, 95% CI [0.25, 2.24]. p=0.015). The difference between active treatment and sham treatment in subjects with lower pain sensitivity was not significant (mean difference -0.45, 95% CI [-1.48. 0.58], p=0.393),
Image /page/12/Figure/5 description: The image is a bar graph comparing PGIC scores at 3 months for sham and active treatments across different populations. The x-axis represents the ITT population, lower pain sensitivity, and higher pain sensitivity. The y-axis represents the PGIC score at 3 months, ranging from 0 to 5. For the ITT population, the sham group has a score of 3.14 (n=57) and the active group has a score of 3.54 (n=62).
Figure 1. Comparison of PGIC in ITT population and in pain sensitivity subgroups
Secondary Effectiveness Endpoints
13
Table 5 shows the 3-month LS mean change scores for the pre-specified secondary effectiveness measures in the ITT population. Negative values indicate improvement. All seven effectiveness measures showed significant within-group (i.e., baseline to 3-months) improvements for active treatment compared to 4 of 7 for sham treatment. The within-group improvements for active treatment were numerically greater than sham treatment for all measures except PCS. The treatment group differences were significant in the FIQR Total Score (-8.58, 95% CI [-15.44, -1.72], p=0.015), BPI Interference (-0.86, 95% CI [-1.64. -0.08], p=0.031), PDQ (-2.21. 95% CI [-3.90, -0.15], p=0.027) and PDI (-4.90, 95% CI [-9.65, -0.14], p=0.044).
| Measure | N | Mean | SE | Diff | Treatment Comparison
(Active - Sham)
95% CI | p-value |
|------------------|------------------|---------|------|-------|---------------------------------------------------|---------|
| | FIQR Total Score | | | | | |
| Sham | 57 | -5.07‡ | 2.53 | | | |
| Active | 62 | -13.65‡ | 2.23 | -8.58 | -15.44, -1.72 | 0.015 |
| FIQR Pain Item* | | | | | | |
| Sham | 57 | -1.01‡ | 0.26 | | | |
| Active | 62 | -1.72‡ | 0.24 | -0.71 | -1.44, 0.02 | 0.055 |
| BPI Severity | | | | | | |
| Sham | 57 | -0.80‡ | 0.23 | | | |
| Active | 62 | -1.28‡ | 0.22 | -0.48 | -1.13, 0.17 | 0.144 |
| BPI Interference | | | | | | |
| Sham | 57 | -0.95‡ | 0.28 | | | |
| Active | 62 | -1.81‡ | 0.26 | -0.86 | -1.64, -0.08 | 0.031 |
| PDQ | | | | | | |
| Sham | 57 | 0.13 | 0.71 | | | |
| Active | 62 | -2.08‡ | 0.65 | -2.21 | -3.90, -0.15 | 0.027 |
| PDI | | | | | | |
| Sham | 57 | -2.40 | 1.76 | | | |
| Active | 62 | -7.30‡ | 1.54 | -4.90 | -9.65, -0.14 | 0.044 |
| HADS | | | | | | |
| Sham | 57 | -0.88 | 0.71 | | | |
| Active | 62 | -2.42‡ | 0.66 | -1.54 | -3.52, 0.43 | 0.124 |
| PCS | | | | | | |
| Sham | 57 | -4.37‡ | 1.25 | | | |
| Active | 62 | 3.66‡ | 1.16 | 0.70 | -2.73, 4.14 | 0.684 |
Table 5. LS mean changes in secondary effectiveness measures from baseline to 3months for the ITT population.
sample t-test.
Table 6 shows the 3-month LS mean change scores for the secondary effectiveness measures in the subgroup with high pain sensitivity. Negative values indicate improvement. Six of 7 effectiveness measures showed significant within-group (i.e., baseline to 3-months) improvements for active treatment compared to 1 of 7 for sham treatment. The within-group improvements for active treatment were numerically greater than sham treatment for all measures. The treatment group differences were significant for FIQR Total Score ( (0(4) p=0.031). FIOR Pain Item ( 0014) 95% CI 95% CI | 同(写) (DX4) p=0.003).
14
| BPI Severity | (b)(4) 95% CI | (b)(4) p=0.035) and PDQ | (b)(6) 95% CI | |
|---|---|---|---|---|
| p=0.018). |
| | | | | | Treatment Comparison
(Active - Sham) | | |
|------------------|----|---------|------|--------|-----------------------------------------|---------|--|
| Measure | N | Mean | SE | Diff | 95% CI | p-value | |
| FIQR Total Score | | | | | | | |
| Sham | 30 | -2.66 | 3.96 | | | | |
| Active | 30 | -13.97‡ | 3.24 | -11.31 | -21.55, -1.07 | 0.031 | |
| FIQR Pain Item* | | | | | | | |
| Sham | 30 | -0.69 | 0.38 | | | | |
| Active | 30 | -2.25‡ | 0.36 | -1.57 | -2.59, -0.54 | 0.003 | |
| BPI Severity | | | | | | | |
| Sham | 30 | -0.31 | 0.34 | | | | |
| Active | 30 | -1.30‡ | 0.31 | -0.99 | -1.90, -0.07 | 0.035 | |
| BPI Interference | | | | | | | |
| Sham | 30 | -0.78 | 0.40 | | | | |
| Active | 30 | -1.86‡ | 0.37 | -1.09 | -2.18, 0.00 | 0.051 | |
| PDQ | | | | | | | |
| Sham | 30 | 0.68 | 1.00 | | | | |
| Active | 30 | -2.68† | 0.95 | -3.35 | -6.13, -0.58 | 0.018 | |
| PDI | | | | | | | |
| Sham | 30 | -0.99 | 2.62 | | | | |
| Active | 30 | -7.35† | 2.26 | -6.37 | -13.16, 0.43 | 0.066 | |
| HADS | | | | | | | |
| Sham | 30 | -0.90 | 1.01 | | | | |
| Active | 30 | -1.90 | 0.97 | -0.99 | -3.74, 1.75 | 0.474 | |
| PCS | | | | | | | |
| Sham | 30 | -3.78† | 1.76 | | | | |
Table 6. LS mean changes in secondary effectiveness measures from baseline to 3-months in the subgroup with high pain sensitivity
Table 7 shows 3-month LS mean change scores of the 21 items comprising the FIQR Total Score in the ITT population. FIOR is a comprehensive health related Quality of Life (OoL) assessment specifically designed for fibromyalgia. It captures pain, somatic symptoms, activities of daily living and overall disease impact. The purpose of this post-hoc analysis is to determine if active treatment broadly improves fibromyalgia symptoms or disproportionally impacts certain symptoms. For active treatment, 19 of 21 symptoms exhibited a significant improvement from baseline to 3-months compared to 5 of 21 for sham treatment. Similarly, 16 of 21 symptoms decreased by at least 1 point compared to 2 of 21 for sham treatment. The active group exhibited significantly better improvement than sham for 9 of the 21 symptoms.
Table 7. LS mean changes in symptoms of fibromyalgia from baseline to 3-months in the ITT population
| FIQR Item (11-point NRS) | Sham | Active | Treatment Comparison
(Active - Sham)
Difference | p-value |
|--------------------------|---------------|---------------|-------------------------------------------------------|---------|
| Pain | -1.01 (0.26)‡ | -1.72 (0.24)‡ | -0.71 (0.36) | 0.055 |
107149
15
| Sleep | -0.55 (0.42) | -1.56 (0.37)‡ | -1.01 (0.58) | 0.080 |
|---|---|---|---|---|
| Fatigue | -0.06 (0.35) | -1.23 (0.31)‡ | -1.17 (0.48) | 0.016 |
| Sensitivity | -1.12 (0.41)† | -1.06 (0.38)† | 0.06 (0.57) | 0.917 |
| Tenderness | -0.21 (0.36) | -0.64 (0.30)† | -0.43 (0.48) | 0.373 |
| Stiffness | -0.60 (0.32) | -1.24 (0.29)‡ | -0.64 (0.44) | 0.150 |
| Balance | -0.35 (0.34) | -1.32 (0.30)‡ | -0.97 (0.46) | 0.039 |
| Anxiety | -0.56 (0.37) | -0.41 (0.31) | 0.14 (0.49) | 0.77 |
| Depression | -0.16 (0.37) | -0.69 (0.31)† | -0.53 (0.50) | 0.294 |
| Memory | -0.71 (0.36)† | -0.85 (0.31)† | -0.13 (0.48) | 0.785 |
| Goals disrupted by fibromyalgia | -0.72 (0.40) | -1.89 (0.33)‡ | -1.16 (0.53) | 0.03 |
| Overwhelmed by symptoms | -0.86 (0.39)† | -2.23 (0.34)‡ | -1.37 (0.52) | 0.010 |
| Ability to walk | -0.76 (0.37)† | -1.22 (0.35)† | -0.46 (0.52) | 0.379 |
| Ability to climb stairs | -0.08 (0.35) | -1.05 (0.33)† | -0.98 (0.49) | 0.050 |
| Ability to clean floors | -0.37 (0.43) | -1.32 (0.39)† | -0.96 (0.59) | 0.108 |
| Ability to shop for groceries | -0.11 (0.38) | -1.15 (0.34)† | -1.05 (0.52) | 0.049 |
| Ability to prepare meal | -0.42 (0.36) | -1.29 (0.31)‡ | -0.87 (0.48) | 0.072 |
| Ability to comb hair | -0.21 (0.31) | -0.51 (0.29) | -0.30 (0.43) | 0.484 |
| Ability to change bed sheets | -0.23 (0.37) | -1.37 (0.32)‡ | -1.14 (0.50) | 0.025 |
| Ability to carry bag of groceries | -0.31 (0.37) | -1.43 (0.33)‡ | -1.12 (0.50) | 0.029 |
| Ability to sit for 45 minutes | -0.31 (0.39) | -1.62 (0.36)† | -1.31 (0.54) | 0.016 |
Responder Analyses
A PGIC responder was defined as a subject with a score ≥ 5 at 3-months, which corresponds to moderately better symptoms, functional abilities and overall health. Forty-three percent (43%) of active treatment subjects in the ITT population and 58% in the higher pain sensitivity subgroup were PGIC responders (Table 8). The difference in the responder rate between active and sham treatment was 28% (p=0.025) in the higher pain sensitivity subgroup.
| Population | Sham % (SE) | Active % (SE) | Diff (95% CI) | p-value |
|---|---|---|---|---|
| ITT (N=119) | 34.6 (6.3) | 42.4 (6.3) | 7.8 (-9.9, 25.5) | 0.389 |
| Subgroup Analysis (N=119) | ||||
| Higher Pain Sensitivity (n=60) | 30.2 (8.4) | 57.8 (9.2) | 27.7 (3.5, 51.8) | 0.025 |
| Lower Pain Sensitivity (n=59) | 39.1 (9.7) | 28.2 (8.0) | -10.9 (-35.5, 13.7) | 0.385 |
| Table 8. PGIC responder analysis. | ||
|---|---|---|
16
A FIOR responder was defined as a subject that exhibited ≥ 15% reduction in their FIQR Total Score from baseline to 3-months. This threshold corresponds to the minimal clinically important difference. Fifty-seven percent (57%) of active treatment subjects in the ITT population and 58% in the higher pain sensitivity subgroup were FIQR responders (Table 9). The difference in the responder rate between active and sham treatment was 23% in the ITT population (p=0.014) and 30% (p=0.019) in the higher pain sensitivity subgroup.
| Population | Sham % (SE) | Active % (SE) | Diff (95% CI) | p-value |
|---|---|---|---|---|
| ITT (N=119) | 34.0 (6.4) | 56.9 (6.5) | 22.9 (4.7, 41.0) | 0.014 |
| Subgroup Analysis (N=119) | ||||
| Higher Pain Sensitivity (n=60) | 28.1 (8.7) | 57.5 (9.2) | 29.5 (4.8, 54.1) | 0.019 |
| Lower Pain Sensitivity (n=59) | 40.3 (9.7) | 55.8 (9.0) | 15.5 (-10.4, 41.5) | 0.240 |
| Table 9. FIQR Total Score responder analysis. | ||||
|---|---|---|---|---|
| -- | ----------------------------------------------- | -- | -- | -- |
A pain intensity responder was defined as a subject that exhibited a > 30% or > 50% reduction in their pain rating (FIOR pain item) from baseline to 3-months. The 30% cutoff represents moderate improvement in pain and the > 50% cutoff represents a substantial improvement in pain. Table 10 shows the pain intensity responder rates. Forty-six percent (46%) of active treatment subjects in the ITT population and 60% in the higher pain sensitivity subgroup were responders at the > 30% level or moderate improvement. Substantial improvement (≥ 50% reduction) was exhibited by 27% of subjects in the ITT population and 43% in the higher pain sensitivity subgroup. The moderate improvement responder rate was 17% greater for active treatment compared to sham (p=0.074). In the higher pain sensitivity subgroup, active treatment was greater than sham by 42% (p=0.001). A similar pattern was observed for substantial improvement responder rates.
| Table 10. Pain intensity responder analysis. | ||||
|---|---|---|---|---|
| -- | -- | -- | ---------------------------------------------- | -- |
| Population | Sham % (SE) | Active % (SE) | Diff (95% CI) | p-value |
|---|---|---|---|---|
| ≥ 30% Reduction in pain intensity from baseline to 3-months (moderate improvement) | ||||
| ITT (N=119) | 29.3 (6.2) | 45.8 (6.7) | 16.5 (-1.6, 34.6) | 0.074 |
| Subgroup Analysis (N=119) | ||||
| Higher Pain Sensitivity (n=60) | 17.5 (7.6) | 59.5 (9.6) | 41.9 (18.6, 65.3) |