Critically Ill, Post Intensive Care Syndrome
Conditions
Brief summary
Objective: Test the ability of vibration to produce physiologic, biochemical, and anatomic changes consistent with exercise that would help prevent the development of muscle weakness that occurs when patients are immobile for long periods of time.
Detailed description
During critical illness, patients who are immobilized for more than a few days develop severe muscle and nerve weakness despite receiving full supportive care, which may include physical therapy. In patients requiring mechanical ventilation (a device that breaths for them) for longer than 7 days, the incidence of ICU-acquired weakness is reported to be between 25% and 60%. Such weakness may contribute to increased duration of mechanical ventilation, increased length of stay in the ICU and hospital, and poor quality of life among survivors. This is part of the newly recognized Post Intensive Care Syndrome (PICS). Moreover, patients who are transferred from the ICU to a high-dependency unit (HDU), intensive therapy unit (ITU), post-operative therapy or outpatient ambulatory care need to be mobile as well as awake for any physical therapy. Patients affected by sepsis (severe blood stream infections), osteoarthritis, spinal cord injury, stroke, multiple sclerosis, cerebral palsy, cancer, and other illnesses suffer muscle loss and weakness. Early mobilization (EM) has demonstrated the ability to significantly reduce the detrimental effects of prolonged immobilization such as polyneuropathy and myopathy (nerve damage and muscle weakness), which in turn reduces the time patients spend on mechanical ventilation and the overall length of hospital stay. EM treatments include intense physical therapy, cycle ergometry, transcutaneous electrical muscle stimulation (TEMS) and continuous lateral rotational therapy (CLRT). However, carrying out intense physical therapy using therapists is impractical (especially at smaller hospitals) and cannot be implemented in heavily sedated patients (patients who cannot cooperate). Evidence suggests that vibration may be capable of producing adequate muscle contraction via muscle-spinal loops that may be sufficient to reduce or prevent nerve damage and muscle weakness caused by prolonged immobilization thus serving as an effective treatment making patients stronger when they leave the ICU. The purpose of this study is to test a prototype vibration device and strategy on its ability to exercise large muscle groups, increase muscle blood flow, and increase circulating levels of blood chemicals associated with exercise/activity. The study will be used to find optimal vibration frequencies that provide maximal evidence of associated muscle activity. Eventually the investigators hope to see a vibration device capable of delivering a more effective therapy compared to the smaller gains derived from traditional measures of physical therapy in critically ill patients such as TEMS, CLRT and cycle ergometry to patients. The vibration device may directly benefit the patient in terms of health, length of stay and reduced re-admission, hospital staff in terms of productivity (i.e., through reduction in nursing effort) and the hospital in terms of reduced cost and return on investment. Its value is also envisioned in many other populations of immobilized acutely ill and injured patients as well as those with chronic conditions. Originally registered as a single record, this registration has been simplified to clarify the outcomes measured from the work with healthy volunteers. A new registration which will include the relevant outcomes for the trial part that will enroll hospitalized participants will be registered prior to their enrollment. The current registration will remain open until it is certain that no additional modifications of the device are required to go through a new round of iterative testing with healthy volunteers. While the total number of participants to be enrolled is larger than some early feasibility trials, the testing is done in small iterative batches to determine whether additional design changes are required. Each of these is generally less than 10 individuals.
Interventions
DIAGNOSTIC_TESTTherapeutic Vibration Device
The Therapeutic Vibration Device is capable of applying force through the axial skeletal spine, through bidirectional compression loading (or prestressing) between the shoulder and the plantar surfaces of the feet. It is placed around the body like a mobile frame so that the applied vibration can affect the whole body. The vibration actuators (drivers) are mobile and can vary in size, frequency response, and force. The design minimizes the possibility of mechanical interference for ventilated/intubated patients.
Sponsors
University of Michigan
Study design
Allocation
NA
Intervention model
SINGLE_GROUP
Primary purpose
PREVENTION
Masking
NONE
Eligibility
Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
Yes
Exclusion criteria
1. Known pregnancy 2. Prisoner
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Change in Regional Hemoglobin Oxygen Saturation | 10 minutes | Change in tissue regional hemoglobin oxygen saturation (rSO2) using near infrared spectroscopy of the thighs,calf, and biceps Baseline measurements were taken for 1 minute and vibration period was for 10 minutes. The mean value of rSO2 for 1 minute preceding vibration was computed as the baseline value. For the data collected during vibration, a moving average peak analysis for every 1 minute for 10 minutes of rSO2 data was carried out. The maximum value of the moving average was selected as the mean value of vibration. The moving average peak analysis was independently conducted for all three measurements from GL, RF and BB. |
| VO2 and VCO2 | baseline and during device use (10 minutes) | Oxygen consumption using a VO2 monitor and mask For the baseline data, a mean of 3 minutes of the segment preceding vibration was computed. For the data collected during vibration, a moving average peak analysis for every 3 minutes for 10 minutes of VO2, VCO2 data was carried out. The maximum value of the moving average was selected as the mean value. This methodology of segment extraction precluded the possibility of picking up short transient changes in metabolic data and helped ensure selection of steady set of values of metabolic variables which estimated the true response of the participant. |
| Energy Expenditure | 10 minutes | For the baseline data, a mean of 3 minutes of the segment preceding vibration was computed. For the data collected during vibration, a moving average peak analysis for every 3 minutes for 10 minutes of EE data was carried out. The maximum value of the moving average was selected as the mean value. This methodology of segment extraction precluded the possibility of picking up short transient changes in metabolic data and helped ensure selection of steady set of values of metabolic variables which estimated the true response of the participant. |
| Minute Variation | 10 minutes | For the baseline data, a mean of 3 minutes of the segment preceding vibration was computed. For the data collected during vibration, a moving average peak analysis for every 3 minutes for 10 minutes of data was carried out. The maximum value of the moving average was selected as the mean value. This methodology of segment extraction precluded the possibility of picking up short transient changes in metabolic data and helped ensure selection of steady set of values of metabolic variables which estimated the true response of the participant. |
| Tidal Volume | 10 minutes | For the baseline data, a mean of 3 minutes of the segment preceding vibration was computed. For the data collected during vibration, a moving average peak analysis for every 3 minutes for 10 minutes of data was carried out. The maximum value of the moving average was selected as the mean value. This methodology of segment extraction precluded the possibility of picking up short transient changes in metabolic data and helped ensure selection of steady set of values of metabolic variables which estimated the true response of the participant. |
| EMG | baseline and during intervention (not exceeding 1 minute) | Simultaneous multi-frequency synchronous excitation was the stimulus, using 15 Hz at shoulders and 25 Hz at feet. Baseline EMG data were recorded prior to commencement of vibration; a 1 second segment was extracted for post processing. For computing muscle activation during vibration, a 10 second EMG segment was extracted after 1 minute of start of the vibration. Extracted signals were filtered to remove artifacts; similar filtering procedures were carried out for EMG signals recorded during MVIC tests and baseline recording. The root-mean square values of EMG signals of vibration and MVIC were calculated. Normalization to MVIC followed (Vibration EMGRMS)/(MVIC EMGRMS) × 100. Bias calculated using (Filtered EMGRMS @ baseline)/(Unfiltered EMGRMS @ baseline); bias-corrected EMG during vibration computed using (Vibration EMGRMS /Bias). Therefore each muscle site has only 1 reported value, representative of the combined effect of multi-frequency excitation provided at shoulders and feet. |
Countries
United States
Participant flow
Recruitment details
Prior to actual testing for the outcomes listed in this registration, 8 participants were recruited and consented for tuning and characterizing the device prior to structured testing.
Pre-assignment details
6 participants consented for testing the device did not come in for their first appointment.
Participants by arm
| Arm | Count |
|---|---|
| Healthy Volunteers (Iterative Device Development) This phase recruited healthy volunteers who were be vibrated with the prototype device using various vibration frequencies to determine which frequency produces the optimal physiologic response. Physiologic responses were determined with a number of devices capable of measuring such things as tissue oxygenation, oxygen consumption, and muscle activity. Volunteers were randomized to receive alternating 5 minute episodes of various vibration frequencies.
Therapeutic Vibration Device: The Therapeutic Vibration Device is capable of applying force through the axial skeletal spine, through bidirectional compression loading (or prestressing) between the shoulder and the plantar surfaces of the feet. It is placed around the body like a mobile frame so that the applied vibration can affect the whole body. The vibration actuators (drivers) are mobile and can vary in size, frequency response, and force. The design minimizes the possibility of mechanical interference for ventilated/intubated patients. | 19 |
| Total | 19 |
Withdrawals & dropouts
| Period | Reason | FG000 |
|---|---|---|
| Overall Study | Withdrawal by Subject | 3 |
Baseline characteristics
| Characteristic | Healthy Volunteers (Iterative Device Development) | — |
|---|---|---|
| Age, Continuous | 51.7 years STANDARD_DEVIATION 19.5 | — |
| Race and Ethnicity Not Collected | — | — Participants |
| Region of Enrollment United States | 19 Participants | — |
| Sex: Female, Male Female | 9 Participants | — |
| Sex: Female, Male Male | 10 Participants | — |
Adverse events
| Event type | EG000 affected / at risk |
|---|---|
| deaths Total, all-cause mortality | 0 / 22 |
| other Total, other adverse events | 3 / 22 |
| serious Total, serious adverse events | 0 / 22 |
Outcome results
Primary
Change in Regional Hemoglobin Oxygen Saturation
Change in tissue regional hemoglobin oxygen saturation (rSO2) using near infrared spectroscopy of the thighs,calf, and biceps Baseline measurements were taken for 1 minute and vibration period was for 10 minutes. The mean value of rSO2 for 1 minute preceding vibration was computed as the baseline value. For the data collected during vibration, a moving average peak analysis for every 1 minute for 10 minutes of rSO2 data was carried out. The maximum value of the moving average was selected as the mean value of vibration. The moving average peak analysis was independently conducted for all three measurements from GL, RF and BB.
Time frame: 10 minutes
Population: One dataset was discarded due to poor data quality due to instrumentation issues encountered during testing.
| Arm | Measure | Group | Value (MEAN) | Dispersion |
|---|---|---|---|---|
| Healthy Volunteers (1st Iteration of Device Development) | Change in Regional Hemoglobin Oxygen Saturation | Thighs (baseline) | 75.65 percent of oxygenation | Standard Error 0.98 |
| Healthy Volunteers (1st Iteration of Device Development) | Change in Regional Hemoglobin Oxygen Saturation | Thighs (during stimulation | 77.01 percent of oxygenation | Standard Error 1.09 |
| Healthy Volunteers (1st Iteration of Device Development) | Change in Regional Hemoglobin Oxygen Saturation | Calf muscle (baseline) | 73.83 percent of oxygenation | Standard Error 0.93 |
| Healthy Volunteers (1st Iteration of Device Development) | Change in Regional Hemoglobin Oxygen Saturation | Calf (during stimulation) | 77.09 percent of oxygenation | Standard Error 1.13 |
| Healthy Volunteers (1st Iteration of Device Development) | Change in Regional Hemoglobin Oxygen Saturation | Biceps (baseline) | 72.23 percent of oxygenation | Standard Error 1.27 |
| Healthy Volunteers (1st Iteration of Device Development) | Change in Regional Hemoglobin Oxygen Saturation | Biceps (during stimulation) | 74.15 percent of oxygenation | Standard Error 1.4 |
Comparison: This presents the p value for the thigh comparison baseline to during stimulationp-value: <0.0001ANOVA
Comparison: This presents the p value for the calf comparisons between baseline and during stimulation.p-value: <0.0001ANOVA
Comparison: This statistical analysis presents the data for the biceps comparisons between baseline and during stimulation.p-value: <0.001ANOVA
Primary
EMG
Simultaneous multi-frequency synchronous excitation was the stimulus, using 15 Hz at shoulders and 25 Hz at feet. Baseline EMG data were recorded prior to commencement of vibration; a 1 second segment was extracted for post processing. For computing muscle activation during vibration, a 10 second EMG segment was extracted after 1 minute of start of the vibration. Extracted signals were filtered to remove artifacts; similar filtering procedures were carried out for EMG signals recorded during MVIC tests and baseline recording. The root-mean square values of EMG signals of vibration and MVIC were calculated. Normalization to MVIC followed (Vibration EMGRMS)/(MVIC EMGRMS) × 100. Bias calculated using (Filtered EMGRMS @ baseline)/(Unfiltered EMGRMS @ baseline); bias-corrected EMG during vibration computed using (Vibration EMGRMS /Bias). Therefore each muscle site has only 1 reported value, representative of the combined effect of multi-frequency excitation provided at shoulders and feet.
Time frame: baseline and during intervention (not exceeding 1 minute)
| Arm | Measure | Group | Value (MEAN) | Dispersion |
|---|---|---|---|---|
| Healthy Volunteers (1st Iteration of Device Development) | EMG | Baseline bellies of soleus (SO) | 13.10 percentage of MVC | Standard Error 1.72 |
| Healthy Volunteers (1st Iteration of Device Development) | EMG | During vibration bellies of soleus (SO) | 62.67 percentage of MVC | Standard Error 17.42 |
| Healthy Volunteers (1st Iteration of Device Development) | EMG | Baseline tibialis anterior (TA) | 5.15 percentage of MVC | Standard Error 0.75 |
| Healthy Volunteers (1st Iteration of Device Development) | EMG | During vibration tibialis anterior (TA) | 11.59 percentage of MVC | Standard Error 2.36 |
| Healthy Volunteers (1st Iteration of Device Development) | EMG | Baseline gastroenemius lateralis (GL) | 3.17 percentage of MVC | Standard Error 0.24 |
| Healthy Volunteers (1st Iteration of Device Development) | EMG | During vibration gastroenemius lateralis (GL) | 7.14 percentage of MVC | Standard Error 1.16 |
| Healthy Volunteers (1st Iteration of Device Development) | EMG | Baseline vastus medialis (VM) | 5.41 percentage of MVC | Standard Error 0.64 |
| Healthy Volunteers (1st Iteration of Device Development) | EMG | During vibration vastus medialis (VM) | 8.52 percentage of MVC | Standard Error 1.25 |
| Healthy Volunteers (1st Iteration of Device Development) | EMG | Baseline vastus lateralis (VL) | 4.33 percentage of MVC | Standard Error 0.52 |
| Healthy Volunteers (1st Iteration of Device Development) | EMG | During vibration vastus lateralis (VL) | 6.57 percentage of MVC | Standard Error 0.73 |
| Healthy Volunteers (1st Iteration of Device Development) | EMG | Baselin rectus femoris (RF) | 4.82 percentage of MVC | Standard Error 1.56 |
| Healthy Volunteers (1st Iteration of Device Development) | EMG | During vibration rectus femoris (RF) | 5.58 percentage of MVC | Standard Error 0.51 |
| Healthy Volunteers (1st Iteration of Device Development) | EMG | Baseline semitendinosus (ST) | 6.31 percentage of MVC | Standard Error 1.97 |
| Healthy Volunteers (1st Iteration of Device Development) | EMG | During vibration semitendinosus (ST) | 6.64 percentage of MVC | Standard Error 0.95 |
| Healthy Volunteers (1st Iteration of Device Development) | EMG | Baseline deltoideus medius | 1.23 percentage of MVC | Standard Error 0.13 |
| Healthy Volunteers (1st Iteration of Device Development) | EMG | During vibration deltoideus medius | 4.25 percentage of MVC | Standard Error 1.05 |
Comparison: Comparison of Vibration to Baseline for Muscle bellies of Soleus (SO)p-value: 0.011ANOVA
Comparison: Comparison of Vibration to Baseline for tibialis anterior (TA)p-value: 0.012ANOVA
Comparison: Comparison of Vibration to Baseline for gastrocnemius lateralis (GL)p-value: 0.003ANOVA
Comparison: Comparison of Vibration to Baseline for vastus medialis (VM)p-value: <0.001ANOVA
Comparison: Comparison of Vibration to Baseline for vastus lateralis (VL)p-value: <0.0001ANOVA
Comparison: Comparison of Vibration to Baseline for rectus femoris (RF)p-value: 0.59ANOVA
Comparison: Comparison of Vibration to Baseline for semitendinosus (ST)p-value: 0.86ANOVA
Comparison: Comparison of Vibration to Baseline for deltoideus mediusp-value: 0.006ANOVA
Primary
Energy Expenditure
For the baseline data, a mean of 3 minutes of the segment preceding vibration was computed. For the data collected during vibration, a moving average peak analysis for every 3 minutes for 10 minutes of EE data was carried out. The maximum value of the moving average was selected as the mean value. This methodology of segment extraction precluded the possibility of picking up short transient changes in metabolic data and helped ensure selection of steady set of values of metabolic variables which estimated the true response of the participant.
Time frame: 10 minutes
Population: Three datasets were discarded due to poor data quality due to instrumentation issues encountered during testing.
| Arm | Measure | Group | Value (MEAN) | Dispersion |
|---|---|---|---|---|
| Healthy Volunteers (1st Iteration of Device Development) | Energy Expenditure | Baseline | 1.12 kcal/minute | Standard Error 0.07 |
| Healthy Volunteers (1st Iteration of Device Development) | Energy Expenditure | Vibration | 1.35 kcal/minute | Standard Error 0.09 |
p-value: <0.0001post-hoc analysis
Primary
Minute Variation
For the baseline data, a mean of 3 minutes of the segment preceding vibration was computed. For the data collected during vibration, a moving average peak analysis for every 3 minutes for 10 minutes of data was carried out. The maximum value of the moving average was selected as the mean value. This methodology of segment extraction precluded the possibility of picking up short transient changes in metabolic data and helped ensure selection of steady set of values of metabolic variables which estimated the true response of the participant.
Time frame: 10 minutes
Population: Three datasets were discarded due to poor data quality due to instrumentation issues encountered during testing.
| Arm | Measure | Group | Value (MEAN) | Dispersion |
|---|---|---|---|---|
| Healthy Volunteers (1st Iteration of Device Development) | Minute Variation | Baseline | 7.09 liters/minute | Standard Error 0.4 |
| Healthy Volunteers (1st Iteration of Device Development) | Minute Variation | Vibration | 8.50 liters/minute | Standard Error 0.55 |
p-value: <0.0001post-hoc analysis
Primary
Tidal Volume
For the baseline data, a mean of 3 minutes of the segment preceding vibration was computed. For the data collected during vibration, a moving average peak analysis for every 3 minutes for 10 minutes of data was carried out. The maximum value of the moving average was selected as the mean value. This methodology of segment extraction precluded the possibility of picking up short transient changes in metabolic data and helped ensure selection of steady set of values of metabolic variables which estimated the true response of the participant.
Time frame: 10 minutes
Population: Three datasets were discarded due to poor data quality due to instrumentation issues encountered during testing.
| Arm | Measure | Group | Value (MEAN) | Dispersion |
|---|---|---|---|---|
| Healthy Volunteers (1st Iteration of Device Development) | Tidal Volume | Baseline | 0.55 liters | Standard Error 0.04 |
| Healthy Volunteers (1st Iteration of Device Development) | Tidal Volume | Vibration | 0.68 liters | Standard Error 0.09 |
p-value: 0.094post-hoc analysis
Primary
VO2 and VCO2
Oxygen consumption using a VO2 monitor and mask For the baseline data, a mean of 3 minutes of the segment preceding vibration was computed. For the data collected during vibration, a moving average peak analysis for every 3 minutes for 10 minutes of VO2, VCO2 data was carried out. The maximum value of the moving average was selected as the mean value. This methodology of segment extraction precluded the possibility of picking up short transient changes in metabolic data and helped ensure selection of steady set of values of metabolic variables which estimated the true response of the participant.
Time frame: baseline and during device use (10 minutes)
Population: Three datasets were discarded due to poor data quality due to instrumentation issues encountered during testing.
| Arm | Measure | Group | Value (MEAN) | Dispersion |
|---|---|---|---|---|
| Healthy Volunteers (1st Iteration of Device Development) | VO2 and VCO2 | VO2 (baseline) | 3.19 ml/(kg*min) | Standard Error 0.16 |
| Healthy Volunteers (1st Iteration of Device Development) | VO2 and VCO2 | VO2 (during stimulation) | 3.84 ml/(kg*min) | Standard Error 0.22 |
| Healthy Volunteers (1st Iteration of Device Development) | VO2 and VCO2 | VCO2 (baseline) | 2.54 ml/(kg*min) | Standard Error 0.14 |
| Healthy Volunteers (1st Iteration of Device Development) | VO2 and VCO2 | VCO2 (during stimulation) | 3.07 ml/(kg*min) | Standard Error 0.18 |
Comparison: The P value below applies to the VO2p-value: <0.0001post-hoc analysis
Comparison: This p value applies to the VCO2p-value: <0.001post-hoc analysis