WO2009046366A1 - Method and apparatus for pressure ulcer prevention and treatment - Google Patents

Abstract

Disclosed are methods and system of using wireless neuromuscular electrical stimulation of appropriate muscles to shift immobile patients from any posture. The methods and apparatus are extended to any immobile person, be it due to spinal cord injury, stroke, dementia, or from any other cause. Accordingly, generated pressure ulcers can be avoided or treated in any anatomic locations of the body based on the electrical stimulations of effective muscles relative to said location. Neuromuscular electrical stimulations of muscles would enable the immobile persons to shift their weight. Full body pressure ulcer prevention protocols using plurality of microstimulators that are implanted at plurality of target locations within the patient's body i.e. locations that are particularly at risk.

Description

Method and Apparatus for Pressure Ulcer Prevention and

Treatment

Cross Reference to Related Applications

[0001] This patent application claims the benefit of the filing date of US provisional application Serial No. 60/977,437, filed October 4, 2007, entitled "Method and Apparatus for Pressure Ulcer Prevention and Treatment" the contents of which are incorporated herein by reference.

Background of the Invention

[0002] Field of the Invention: This application relates generally to devices and methods to activate muscles of immobile patients in order to shift their weight using neuromuscular electrical stimulation (NMES).

[0003] General Background and State of the Art: Pressure ulcers (PUs) are common and debilitating wounds that arise when immobilized patients cannot shift their weight. Able-bodied people do not get PUs because they can voluntarily contract their muscles, thereby shifting their weight while activating trophic mechanisms that maintain muscle bulk, strength and circulation. Pressure ulcers (PUs) are a debilitating pathology resulting from pressure and shear in the soft tissues of immobilized patients. Blood vessels become occluded and the soft tissues they supply necrose.

[0004] Approximately 30% of immobile patients develop PUs. Many preventive or therapeutic practices and devices continue in use, but to little avail. Despite a team of highly trained caregivers and the most intensive therapies available, the death of Christopher Reeve from PU-related complications highlights the need for novel approaches to successful PU prevention (PUP) and treatment. For prevention the mainstay is tissue load reduction, typically by regular weight-shifting together with a variety of passive cushions and pneumatic devices. It is recommended that immobile patients change position every 2 hours when recumbent and every 15-20 min when seated. This high demand often results in poor compliance, especially when patients do not feel pain or discomfort, or are ineffective at independently shifting their weight. Of PU treatment modalities, the conservative mainstay is prolonged passive load reduction with medical care, but extensive surgical repair is often necessary. Even with aggressive surgery, at high costs per incident, recurrence rates can reach 61 % within the first year of repair. Age and pathology matched patient trials have shown that hospital stays increase 3-5 fold in patients suffering from PUs, incurring significant additional expense.

[0005] In addition to elderly patients who lose mobility from stroke, dementia, Parkinson's disease, etc., one group particularly prone to PUs is those with spinal cord injury (SCI). The prevalence of SCI in the US has been estimated at 253,000 (225,000-296,000, June 2006), with an incidence of 1 1 ,000-12,000 new injuries each year (40 per million US population). SCI most commonly results in paralysis, as well as repeated and serious complications including PUs, incontinence, pneumonia, etc. The incidence of PUs in SCI has been estimated by the Model Spinal Cord Injury Systems centers at one third (33.5%) during initial acute care and rehabilitation, and at 30% thereafter. Prevalence has been estimated at a similar percentage (Fuhrer, et al. found that 33% of a community-based sample of 140 randomly selected SCI persons had a PU on examination). PU severity is most commonly staged according to the National Pressure Ulcer Advisory Panel system, originally put forward in 1989. Throughout its evolution, it always remained a 4-stage system ("Stages I-IV"), until two additional stages were added in 2007, to include "Suspected Deep Tissue Injury" and "Unstageable" categories. Extensive reconstructive surgery is usually required for Stage III PUs (full thickness tissue loss), and invariably so for Stage IV PUs (exposed bone, tendon or muscle).

[0006] Flap reconstruction to provide well-vascularized, bulky tissue to repair PUs over bony prominences was pioneered by Davis in the 1930s. Since 1970, gluteal flaps have been widely used to this purpose (technique originally described by Ger 27,51 ). Although still the best option available, post-operative recurrence rates are high as the flap tissue in SCI patients is not truly healthy, thick, vascularized and resistant to PUs, as in healthy subjects. Reported recurrence rates vary widely. In 1992 Disa, et al. reported on 66 flaps: 61 % of PUs and 69% of patients had recurrences within a mean of 9.3 months, despite 80% having healed at discharge. In 1994 Evans' group found an 82% recurrence rate at the site of surgical repair, and 64% at a new site, in 22 paraplegic patients. In 1997 Foster, et al. reviewed 139 ischial PUs in 1 14 consecutive patients over 16 yrs: on admission prior flap reconstruction had been performed in 60%, and a 17% recurrence rate was noted at a mean follow-up of 10.7 months. In 1998 Kierney's group reviewed 268 PUs operated in 158 patients over a 12 year period with a mean follow-up of 3.7 years. The recurrence rate for ischial flap closures was 21 %; the overall rate for all flaps in SCI patients was 23% (24% for paraplegics; 20% for tetraplegics). In 1999 Tavakoli, et al. also found a higher recurrence rate in paraplegics than tetraplegics (57% vs. 33.3%, n=23, mean follow-up 62 months).

[0007] In 2000 Schryvers' group retrospectively assessed 191 flap repairs of ischial PUs over a 20 year period, finding a recurrence rate of 34% (follow-up 2 mos to 3 yrs). Finally, in 2003 Margara, et al. reviewed 121 ischial PUs repaired over a 15 year period. They reported a recurrence rate of 33% (19/57) over the first 7 years, which was reduced to 9% (6/64) in the subsequent 8 years according to a stricter treatment protocol. The literature is too fragmentary to confirm the suggestion of a gradually improving trend, but it confirms the obvious fact that postoperatively patients will resume sitting on their repaired sites, again occluding capillary circulation and recreating shear conditions.

[0008] Pathophysiology of Pressure Ulcers

[0009] The pathogenesis of PUs derives from a host of compounding etiological factors. Such as pressure over bony prominences. A reciprocal relationship between pressure intensity and pressure duration has been recognized. Just as high pressures over a short time may result in PUs, so can lower pressures over a longer time. Immobility that results in the inability to shift weight are due to the tissues not experiencing periodic relief from the pressures on them. Sustained pressure reduces or occludes capillary circulation, resulting in hypoxia and necrosis of soft tissues.

[0010] Friction and shear can be other factors, because they can exacerbate the soft tissue damage. Tissue that is damaged, atrophied, scarred or infected demonstrates increased susceptibility to pressure. Elderly or immune compromised patients and patients with poor wound healing or collagen-vascular diseases are at greater risk too.

[0011] A lack of sensation, common in many poorly mobile individuals, aggravates the situation further, because these patients may forget to move even to the extent that they can. Insensate tissues are prone to compromised neurotrophic growth and repair mechanisms. Urinary and/or fecal incontinence may also be present in patients who are immobile due to SCI or CNS pathologies. The resultant irritation promotes maceration and infection of the skin, leaving these patients even more susceptible to the effects of pressure. As blood vessels become occluded or narrowed, the soft tissues which they supply necrose (break down and die), as they are starved of nutrients and oxygen, whilst accumulating toxic wastes and metabolites.

[0012] Staging of Pressure Ulcers

[0013] A widely accepted method of staging PUs according to the National PU Advisory Panel are as follows:

[0014] Stage I: Non-blanchable erythema of intact skin. The heralding lesion of skin ulceration.

[0015] Stage II: Partial thickness skin loss involving epidermis and/or dermis. A superficial ulcer that presents clinically as an abrasion, blister or shallow crater.

[0016] Stage III: Full thickness skin loss involving damage or necrosis of subcutaneous tissue which may extend down to, but not through, underlying fascia. The ulcer presents clinically as a deep crater with or without undermining of adjacent tissues.

[0017] Stage IV: Full thickness skin loss with extensive destruction, tissue necrosis or damage to muscles, bone or supporting structures (e.g. tendon, joint, capsule, etc.).

[0018] Current treatments involve: Tissue Load Reduction, Nursing Care & Allied Therapies, Treatment of Complications & Supportive Care, Wound Care & Surgical Interventions, and Adjuvant Therapies. Treatment of PUs usually involve months of passive tissue-load reduction and commonly hospitalization and surgical intervention. SUMMARY

[0019] Exemplary embodiments of the pressure ulcers (PUs) treatment and alleviation systems and methods include activating muscles that shift the position of the soft tissues that are at risk of pressure due to immobility; to achieve mechanical activation of said muscles; provide fully implanted means to deliver electrical stimulation to said nerves and muscles; to provide wireless means for transmitting electrical power and control signals from outside the body to said implanted electrical stimulators; alleviating pressure ulcers by one or more wireless miniature device that can be implanted in or near target muscles without the requiring leads for electrodes; to implant a miniature device through a tube, needle or any other injection device, in or near a target muscle to prevent or alleviate pressure ulcers.

[0020] Additional exemplary embodiments of the pressure ulcers (PUs) treatment and alleviation systems and methods include selectively controlling multiple implanted miniature devices to produce a variety of patterns of muscle contraction that will be effective and well tolerated by each patient; and triggering the appropriate stimulation of muscles for a given assumed posture by utilizing position sensors and to customize effective muscle stimulations patterns and/or programs person to person through fitting assessments to constitute effective pressure-relief programs.

[0021] It is understood that other embodiments of the pressure ulcers (PUs) treatment and alleviation systems and methods will become readily apparent to those skilled in the art from the following detailed description, wherein it is shown and described only exemplary embodiments by way of illustration. As will be realized, the pressure ulcers (PUs) treatment and alleviation systems and methods are capable of other and different embodiments and its several details are capable of modification in various other respects. Accordingly, the drawings and detailed description are to be regarded as illustrative in nature and not as restrictive.

BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figures 1 A and 1 B show incremental loading curves for models of different somatotypes.

DETAILED DESCRIPTION

[0023] The detailed description set forth below is intended as a description of exemplary embodiments of the pressure ulcer prevention (PUP) or treatment system and method and is not intended to represent the only embodiments in which the pressure ulcer prevention or treatment system and method can be practiced. The term "exemplary" used throughout this description means "serving as an example, instance, or illustration," and should not necessarily be construed as preferred or advantageous over other embodiments. The detailed description includes specific details for the purpose of providing a thorough understanding of the pressure ulcer prevention or treatment system and method. However, it will be apparent to those skilled in the art that the pressure ulcer prevention (PUP) or treatment system and method may be practiced without these specific details. In some instances, well- known structures and devices are shown in block diagram form in order to avoid obscuring the concepts of the pressure ulcer prevention or treatment system and method.

[0024] Exemplary embodiments comprise PUP through neuromuscular electrical stimulation (NMES) using BIONs (wireless, injectable microstimulators), a method and apparatus for using NMES of appropriate muscles to shift immobile patients from any posture, for pressure ulcer prevention and treatment of all anatomic locations. This approach may be utilized by patients with immobility such as due to SCI, stroke, dementia, or from any other cause.

[0025] BIONs are injectable, wireless electrical stimulators that receive power and command signals by inductive coupling from an external transmission coil(s). One or more can be placed in or near various muscles and nerves and selectively activated with precise control of pulse current, duration and timing. BIONs have been demonstrated to be safe and effective in chronic studies in animals and human subjects. Although any NMES stimulator may be suitable for this invention, BIONs are particularly suited because of their small size ( Approximately 2mm diameter x 16mm long), hermetic packaging, stable fixation in connective tissue, and virtually unlimited lifetime. Transmission coil(s) to power and command the implants can be located within the wheelchairs, seats, beds or deep to any other supports that immobile patients require, so avoiding any direct contact with or penetration of the patient's already fragile skin. Using BION microstimulators to electrically stimulate muscles is a useful way to achieve the benefits of NMES for preventing PUs, without the usual limitations associated with wired electrodes as discussed above. The treatment may also be self-administered by patients.

[0026] We present a method and apparatus to facilitate shifting of a subject's weight distribution to prevent PUs using BIONs placed alongside the nerves to appropriate muscles as demanded by the PU risk at hand. This applies to pelvic PUs as well as other anatomical locations.

[0027] Full-body PUP through NMES

[0028] NMES of appropriate muscles, using BIONs stimulator, are capable of redistributing a patient's weight, so as to relieve pressure and prevent PUs. [0029] Tissue health will be improved during NMES due to increased muscle perfusion.

[0030] Wound healing may be improved with increased wound-healing rates and/or decreased wound-healing complications.

[0031] At-risk Anatomical Locations of PUs

[0032] The following locations are the exemplary at risk locations for PUs:

Sacrum 33.3 %

Heel 17.4 %

Ischium 12.0 %

Unclassified 6.4 %

Genitals 4.9 %

Scapula 4.4 %

Foot 4.4 %

Trochanter 3.1 %

Occiput 2.8 %

Malleolus 2.7 %

Elbow 2.6 %

Spinous Process 1.8 %

Knee 1.7 %

Iliac Crest 1.5 %

Rib 1.0 %

Total 100.0 %

[0033] BIONs may be implanted via percutaneous injection or during any concurrent surgery such as flaps for PU reconstruction ("healthy", although atrophic, muscle and skin are brought in to repair the deficient area, and to provide "healthy" tissue over bony prominences, while scars are designed to lie away from these areas). Together with the prophylactic NMES, all patients may receive standard nursing care (including relief of pressure with cushions, regular turnings, etc.).

[0034] NMES with BIONs™ (injectable microstimulators) can be used to wirelessly activate maximal muscle contractions in order to produce skeletal torques and motion, with associated increases in muscle bulk (hypertrophy), strength, and metabolic capacity; hence counteracting the three major etiological factors in PU development which are immobility, soft-tissue atrophy, and hypoxia.

[0035] The BION is a wireless, injectable microminiature stimulator (approximately 2.1 mm diameter x 15.5mm long). This new class of generic neural prosthetic device can deliver precise electrical stimulation pulses to an arbitrary number of nerve and muscle sites. BIONs receive power and digital command signals by RF telemetry from an external transmission coil.

[0036] An exemplary embodiment of the present system is for using neuromuscular electrical stimulation (NMES) of appropriate muscles to shift immobile patients from any posture, for PU prevention (PUP) and treatment of all anatomic locations. This may involve a standard relief protocol for seated paraplegics / quadriplegics, and another for bed-ridden paraplegics / quadriplegics. The full preventive protocols may require 5-7 BIONs per side in paraplegics and 10-15 per side in quadriplegics; however they may be modified or used in any of a variety of partial combinations if any specific sites are particularly at risk.

[0037] The proposed muscle groups which may require activation, and nerves which may require stimulation to achieve such, include, but are not limited to those muscles which may actuate movements antagonistic to the at-risk posture, whether seated, prone, supine, or recumbent posture (on one's side) postures. For example, as shown in Fig. 1 (As also discussed in the attached appendix I), when ones hips are extended in the supine position, gluteus maximus (GM) is a dominant hip extensor, and so the potential exists for GM stimulation to extend and abduct the hip and so roll bed-ridden patients axially. This may relieve both ischial and sacral pressures in supine patients and so may be a beneficial treatment for this group of PU patients. Similarly, as detailed below, other muscle groups may be employed as needed to facilitate pressure relief in any area of the body.

[0038] In a preferred embodiment, one or more microstimulators may be implanted in or near the appropriate nerve as suggested by examples in Tables 1 -4 below. Table 1 - Examples of Stimulation to Overcome At-Risk Areas When Seated

ANATOMIC % OF EXAMPLES OF ACTION (AS DESIRED MAY BE EXAMPLES OF

LOCATION PUS DESIRED BILATERAL OR ALTERNATING SIDE NERVES TO

MUSCLE(S) TO SIDE) STIMULATE (NOT ALL REQD)

Saccrum 33.3 Hamstrings Extend hip lifting sacrum forwards away Sciatic from seat back

Rectus Abdominis Flex hip pulling sacrum forwards away Thoracoabdominal from seat back

Heel 17.4 Quadriceps Extend knee lifting heel off footrest Femoral

Gastrocnemius Plantar flex ankle lifting heel off footrest Tibial

Soleus Plantar flex ankle lifting heel off footrest Tibial

Ischium 12.0 Hamstrings Extend hip lifting ischium up off seat Sciatic

Genitals 4.9 Hamstrings Extend hip lifting genitals up off seat and Sciatic from in between thighs

Scapula 4.4 Rectus Abdominis Flex hip pulling trunk forwards and Thoracoabdominal scapula away from seat back

Foot 4.4 Quadriceps Extend knee lifting heel off footrest Femoral

Elbow 2.6 Biceps (long head) Flex shoulder to lift elbow off arm rest Musculocutaneous

Anterior Deltoid Flex shoulder to lift elbow off arm rest Axillary

Pectoralis Major Flex shoulder to lift elbow off arm rest Pectoral nn.

Spinous 1.8 Rectus Abdominis Flex hip pulling trunk forwards and Thoracoabdominal process scapula away from seat back

Total 80.8 Table 2 - Examples of Stimulation to Overcome At-Risk Areas When Prone

ANATOMIC % OF DESIRED ACTION (AS DESIRED MAY BE EXAMPLES OF LOCATION PUS MUSCLE(S) BILATERAL OR ALTERNATING NERVES TO SIDE TO SIDE) STIMULATE (NOT ALL REQD)

Genitals 4.9 lliacus Flex hip lifting genitalia off bed Femoral (muscular branch)

Psoas Flex hip lifting genitalia off bed Lumbar plexus (L1 -L3)

Rectus Femoris Flex hip lifting genitalia off bed Femoral (muscular branch)

Rectus Abdominis Flex hip lifting genitalia off bed Thoracoabdominal

Foot 4.4 Hamstrings Flex knee lifting foot off bed Sciatic

Knee 1.7 Quadriceps Extend knee alleviating pressure on Femoral knee

Iliac Crest 1.5 lliacus Flex hip lifting pelvis off bed Femoral (muscular branch)

Psoas Flex hip lifting pelvis off bed Lumbar plexus (L1 -L3)

Rectus Femoris Flex hip lifting pelvis off bed Femoral (muscular branch)

Rectus Abdominis Flex hip lifting pelvis off bed Thoracoabdominal

Rib 1.0 lliacus Flex hip, alternating side to side, to Femoral (muscular raise and rock torso branch)

Psoas Flex hip, alternating side to side, to Lumbar plexus (L1 -L3) raise and rock torso

Rectus Femoris Flex hip, alternating side to side, to Femoral (muscular raise and rock torso branch)

Rectus Abdominis Flex hio. alternatinα side to side. to Thoracoabdominal raise and rock torso

Total 13.5

Table 3 - Examples of Stimulation to Overcome At-Risk Areas When Sue

ANATOMIC % OF DESIRED ACTION (AS DESIRED MAY BE EXAMPLES OF LOCATION PUS MUSCLE(S) BILATERAL OR ALTERNATING SIDE NERVES TO TO SIDE) STIMULATE (NOT ALL REQD)

Saccrum 33.3 Gluteus Maximus Extend hip lifting sacrum up off bed Inferior Gluteal Hamstrings Extend hip lifting sacrum up off bed Sciatic Heel 17.4 Quadriceps Extend knee while flexing hip to lift foot Femoral off bed lliacus Flex hip while extending knee to lift foot Femoral (muscular off bed branch)

Psoas Flex hip while extending knee to lift foot Lumbar plexus (L1 - off bed L3)

Rectus Femoris Flex hip while extending knee to lift foot Femoral (muscular off bed branch)

Rectus Abdominis Flex hip while extending knee to lift foot Thoracoabdominal off bed

Scapula 4.4 Rectus Abdominis Flex hip to raise trunk off bed Thoracoabdominal Occiput 2.8 Sternocleidomastoid Flex neck while flexing hip to raise head Accessory off pillow

Rectus Abdominis Flex hip while flexing neck to raise head Thoracoabdominal off pillow

Malleolus 2.7 Gluteus Medius Internal rotation of leg Superior Gluteal Hamstrings Internal rotation of leg Sciatic Elbow 2.6 Biceps (long head) Flex shoulder to lift elbow off bed Musculocutaneous Anterior Deltoid Flex shoulder to lift elbow off bed Axillary

Spinous 1.8 Rectus Abdominis Flex hip to raise trunk off bed (alternating Thoracoabdominal Process with extension; alternate sides)

Erector Spinae Extend trunk (alternating with flexion; Spinal nn. (C1— S5) alternate sides)

Rib 1.0 Rectus Abdominis Flex hip to raise trunk off bed (alternating Thoracoabdominal sides)

Total 66.0

Table 4 - Examples of Stimulation to Overcome At-Risk Areas When Recumbent

ANATOMIC % OF DESIRED ACTION (AS DESIRED MAY BE EXAMPLES OF

LOCATION PUS MUSCLE(S) BILATERAL OR ALTERNATING SIDE NERVES TO TO SIDE) STIMULATE (NOT ALL REQD)

Scapula 4.4 Combination of

Foot 4.4 muscles to extend torso and role into

Trochanter 3.1 supine position,

Malleolus 2.7 followed by regular rocking through

Elbow 2.6 alternating hip

Iliac Crest 1.5 extension with

Rib 1.0 intermittent leg flexion to relieve

Total 19.7 heels and malleoli Said microstimulators are advantageously of a size and shape to be implantable through the lumen of a needle or needle-like insertion tool. In the preferred embodiment, said microstimulators (10) may be approximately 2mm diameter by 16mm long. The function, form and detailed design of such microstimulators, and of said insertion tool, have been described in detail in previous patents and patent applications that may be used in various embodiments are described in US 5,193,539; US 5,193,540; US 5,312,439; US 5,324,316; US 20030233125 and US 1 1/680,363, the contents of which are incorporated by reference in their entirety.

[0039] Each microstimulator may consist of three major elements: electronic subassembly and two electrodes. Each separate microstimulator's electronic subassembly may receive power and individually addressed command signals by inductive coupling from transmission coil(s) located outside the body, advantageously in a pillow, seat cushion, seatback, mattress, clothing, or the like so as to guarantee proximity to the implanted microstimulators. Transmission coil(s) may be energized with a radio frequency electrical current generated by a driver, and modulated according to a pattern that has been loaded into digital memory means contained within the controller. The modulation pattern constitutes a string of command signals. Each command may contain digital data identifying which microstimulator is to generate which electrical stimulation pulses and the timing and intensity required to evoke the desired muscle contraction.

[0040] During the initial implantation, it may be advantageous to test the efficacy of potential sites for the implantation of microstimulators by applying stimulation pulses through conventional electrodes that may or may not be incorporated into or passed temporarily through the instruments used for implantation. One instrument to utilize is the present inventors's own implant (microstimulator) injection device published as US 20030233125 which is incorporated by reference in its entirety. Through this device, practitioner is able to penetrate patient's body while the implant is held within the device, and further test the efficacy of the site before releasing the implant. The modular design and addressability of said microstimulators is advantageous because it permits additional channels of stimulation to be added to the patient at any time without interfering with those channels installed previously.

[0041] After implantation, the design of an effective pattern of stimulation pulses can be accomplished by variously activating the implanted microstimulators via the transmission coil(s). This may be implemented while assessing pressure relief of the at-risk area of interest by visibly noting effective weight-shifting, measuring and analyzing the redistribution of interface pressures, assessing tissue health over time, and any combination of these or other methods (Gluteal area is discussed in the two appendices attached to and integral to this disclosure). In this way standard relief protocols may be designed and tailored to specific patients for specific postures, and these may be programmed into the controller at the fitting session(s).

[0042] To do this the clinician may use specialized software in a personal computer to devise various stimulation patterns and to deliver them to the controller, which formats them for transmission to the implanted microstimulators via the transmission coil(s) and driver(s). When an effective stimulation program has been identified, it may be loaded into the non-volatile memory in the controller so that the patient can use the controller to deliver the stimulation program(s) at home. When the controller is turned on and running a stimulation program, the commands may be sent to the microstimulators inside the patient. This communication requires the patient to be within the magnetic field generated range by transmission coil(s) and driver(s). [0043] During treatment, the patient may initiate an appropriate program as needed by switching it on at the controller; or the appropriate program may be triggered automatically by position sensors worn on the patient which may register each current posture, such as accelerometers or much like commercially available activity monitors. In an exemplary embodiment, the stimulation pattern may be applied automatically and continuously to prevent pressure on the tissues in whichever posture the patient assumes.

[0044] The BION can be implanted next to each of the inferior gluteal neurovascular pedicle and the proximal sciatic nerve (which supply gluteus maximus and the remaining hamstring hip extensors respectively), BION Active Seating (BAS) can be used in seated paralyzed patients to build up gluteal muscle volume and blood circulation, and intermittently may unload the soft tissues under the ischium. Thereby preventing occurrences and recurrences of PUs.

[0045] Neuromuscular electrical stimulation (NMES) via chronically implanted, wireless microstimulators (BIONs™) might be used both to shift the paralyzed subject's weight as well as to build up gluteal muscle volume and blood circulation, in order to prevent occurrences and recurrences of PUs. Gluteus maximus has been considered an actuator for hip extension in trying to relieve seated pressures through NMES. We have modeled the skeletal biomechanics of this system to quantify the value of gluteus maximus relative to other hip extensors (hamstrings), using rigid body mass and muscle moment analyses. Surface stimulation experiments may be used to validate our findings in order to identify promising stimulation sites and strategies for such treatment.

[0046] While we recognize that gluteal stimulation may likely to reduce disuse atrophy and improve circulation, we show that it may neither be required, nor desired, to achieve hip extension by NMES. Instead hamstring stimulation may be required, and it may provide sufficient hip extension to relieve hydrostatic pressures in the soft tissues under the ischium.

[0047] Skeletal Biomechanics

[0048] NMES with BIONs™ can be used to activate maximal muscle contractions, and produce skeletal motion with associated increases in muscle bulk (hypertrophy), strength, and metabolic capacity. Hence it may counteract the three major etiological factors in PU development (immobility, soft-tissue atrophy, and hypoxia).

[0049] The BION Active Seating (BAS) can be used to prevent occurrences and recurrences of PUs in paralyzed patients by separately building up gluteal muscle volume and blood circulation, and intermittently unloading the soft tissues under the ischium when seated.

[0050] We have performed anatomical dissections with 3D CT reconstructions to : 1 ) map the inferior gluteal neurovascular pedicle (to GM) and the proximal sciateic nerve (to HS), and 2) confirmed that BIONs can be placed at these sites reliably with the surgical approach afforded by gluteal rotation flap repair of PUs. Intraneural fascicular dissection of the sciatic nerve has further demonstrated that the 4 HS motor branches all lie superficially and midially within the nerve wher it becomes accessibale as it exits the greater sciatic forarm beneath piriformis. This suggests that a BION at this location can stimulate HS selectively.

[0051] In the exemplary embodiment, we have implanted BIONs next to each target nerve location in 3 subjects, and have been able to activate GM and HS selectively in all cases. [0052] Our approach has the following advantages: 1 ) access is readily available and BION placement is precise, so reducing confounding variables related to optimal electrode location (minimizing charge required to elicit a response and risk of cross- stimulation of nearby nerves); 2) each patient acts as their own intra-patient control; and 3) patients who require PU repair are at especially high-risk of recurrence, so constituting a particularly good group in which to assess the efficacy of a preventive treatment. Furthermore, we are afforded the opportunity for proof-of-concept of the ability to: 1 ) stimulate the HS and GM selectively with BIONs; and 2) achieve the desired weight-shifting, muscle hypertrophy and perfusion benefits.

[0053] Finally we are able to gather anatomical landmark data to plan simple percutaneous approaches for any future full (bilateral) therapy. The latter is particularly important for the GM location, where surface landmarks are not well- described; but also for access to the medial sciatic nerve, for which current sciatic block descriptions are generally directed at complete blockade.

[0054] It has been reported that 69% of PUs are in the pelvic region. Of these, BAS may potentially benefit both sacral and ischial lesions, which constitute 42-54% of all PUs. Of this group 1 1 % are Stage III & IV, where surgery is often (SIII) or always (SIV) the treatment of choice. The gluteal rotation flap is a common reconstructive technique for treating PUs by bringing healthy (although often atrophied) muscle and skin in to repair the necrotic area and provide healthy tissue over bony prominences.

[0055] We now present an exemplary system to identify effective implant locations for the BIONs and stimulation parameters for relieving internal soft tissue pressures. We have modeled the skeletal biomechanics of this system to quantify the value of gluteus maximus (GM) stimulation relative to sciatic extensor (SE) (hamstring) stimulation, using rigid body mass and muscle moment analyses. Surface stimulation experiments were used to validate model predictions. While we recognize that GM stimulation may likely reduce disuse atrophy and improve circulation, we include here that it may not either be required, nor desired, to achieve hip extension in seated subjects. Instead SE stimulation may be required, and is capable of providing sufficient hip extension to relieve hydrostatic pressures in the soft tissues under the ischium. Both GM and SE stimulation can probably be used for muscle conditioning when the patient is non weight bearing. We later in the disclosure present the model of the soft tissue biomechanics of this system and develop a treatment rationale.

[0056] MODELING OF SKELETAL BIOMECHANICS

[0057] RIGID BODY MASS & MUSCLE MOMENT ANALYSES

[0058] A. Rigid Body Mass Analysis

[0059] 1. Methods

[0060] To ascertain the potential effects of NMES on the paralyzed musculoskeletal system, the biomechanics of this system has been reconsidered from first principles using rigid body analysis.

[0061] The biomechanics of the seated human has been reconsidered from first principles using rigid body analysis, to determine the postural effects of isolated hip extension on the paralyzed musculoskeletal system, regardless of how that extension was achieved were identified (MSC. visual Nastran 4D, MSC Software Corp, CA). The body was represented by 8 rigid segments, seated in a rigid seat with footrest: 3 for each lower limb ('thigh', leg' and 'foot'), 1 for the pelvis ('pelvis'), and 1 for the torso, upper limbs, and head & neck ('torso').

[0062] While it is recognized that SCI subjects do not have the same mass proportions of their body segments as able-bodied individuals, we have elected to use able-bodied mass proportions for three reasons: 1 ) the aim is to design a treatment that ultimately can be implemented soon after injury for prophylaxis (before and to avoid atrophic changes); 2) a goal of the treatment is to strengthen and hypertrophy muscles to approach the bulk and strength of able-bodied counterparts; and 3) sufficient torque will be required to overcome, at worst, segment masses approaching those of able-bodied individuals. The dimensions and mass of each segment were therefore apportioned according to the HYBRID III Fiftieth Percentile Crash Test Dummy , which represents an average adult male of 5'9" and 78 kg. Seventy percent of the total mass was apportioned to the body segments (mBody = 55 kg, comprising mPelvis = 9 kg and mTorso = 46 kg), and the remaining 30% of the mass was distributed between the lower limb segments (mLLimb = 1 1.5 kg each, comprising mThigh = 6 kg, mLeg = 4.75 kg and mFoot = 0.75 kg).

[0063] All segments were connected by passive joints, except for the R. hip (the site of "stimulation"), which was actuated with an extension torque motor. The pelvis- torso joint was considered rigid, on the assumption that for the small shifts we are modeling, the lumbo-sacral joint would offer only a limited range of motion (ROM) relative to the remaining joints. Based on the same assumption of limited ROM, the ankles and knees were modeled as frictionless, spherical joints; and the L. hip as a frictionless, revolute joint. The R. hip torque was calculated at 63 Nm by assuming that the knee was initially located near the edge of the seat, that mBody was evenly distributed across the hips, that we wished to raise the body by 30°in 0.5 s, and that it would take 0.2 s to attain this angular velocity, as follows:

τ = r -F where r = t - sin θ , F = m - a , a = v - u and v = _ω • 2Ht t 360°

where T = torque, r = moment arm, F = force, I = displacement arm length (pelvis-knee = 0.6 m), θ = angle of F from I (90o, as we wish to calculate maximum T required), m = mass (0.5^*mBody + mThigh = 33.5 kg), a = acceleration, v = final linear velocity, u = initial linear velocity (0 m/s), t = time (0.2 s), and ω = angular velocity (60 deg/s).

[0064] The resultant torque of 63 Nm that was required for this scenario. This correlates well with experimentally derived torques for hip extension (100-180 Nm are achievable using all extensors).

[0065] 2. Results

[0066] In healthy subjects, surface stimulation of the GM and HS hip extensors is inevitably accompanied by involuntary co-contraction of their antagonists and other agonists. During GM stimulation cross-stimulation of the sciatic nerve (HS) can also occur. Further, accurately measuring postural changes in this system can present its own hurdles, even in healthy subjects. Therefore the model was useful to appreciate the unloading effects of an isolated unilateral hip extensor torque.

[0067] Qualitatively the model behaved as expected intuitively for multiple simulations over a range of static loads, and was validated by comparison with experimental behavior recorded during surface stimulation experiments (see below). It was confirmed that the calculated torque requirement of 63 Nm can provide sufficient unloading for even the extreme scenario proposed, which established this torque as a goal for the subsequent models of specific muscle actions (described below). It also suggested that pressure relief could be achieved not only ipsilaterally, but possibly also contralateral^ to some extent.

[0068] B. Muscle Moment Analysis

[0069] 1. Methods

[0070] Surface stimulation experiments (described below) demonstrated that HS stimulation decreased ischial pressure by 26%, while GM stimulation increased ischial pressure by 16%. The relative hip extensor capabilities of the HS and GM groups were compared. An anatomically accurate model of muscle origin/insertion and wrapping was analyzed in SIMM (Software for Interactive Musculoskeletal Modeling; MusculoGraphics, Santa Rosa, CA) to identify the axis and magnitude of the relative hip torques that would be produced by each type of stimulation. All other muscles were inactivated and the musculoskeletal model was set with thighs neutral at 0°of internal-external rotation and 0°of abduction-adduction.

[0071] Force (N) and moment arm (m) data were derived for each contributory muscle across its full ROM in flexion-extension, to calculate the torque/moment data for each at maximal activation: Torque (Nm) = Moment Arm (m) x Force (N). The moments of all muscles in each group were summed to plot GM total moment and HS total moment vs. ROM in flexion-extension. Similar data was derived for internal- external rotation and abduction-adduction moments in the seated posture.

[0072] 2. Results

[0073] Total hip extensor torque: Comparing the GM component sum and the HS component sum confirms that GM may be the dominant hip extensor during walking, i.e. between heel contact and toe-off of a regular gait cycle, when the hip angle range is approximately from +30°to -10°. When seated at approximately 90° of hip flexion, however, over 80% of the available extensor torque may originate from HS, and not from GM, a proportion that persists over as much as 20°of hip extension / unloading.

[0074] HS stimulation alone may provide ~100Nm of extension torque while only ~60Nm may be required for even extreme unloading (as detailed above). GM stimulation provided only -20 Nm of extension torque in the seated posture, suggesting that even a non-atrophied GM would be insufficient to achieve ischial elevation. The differential effects of GM and HS activation suggested here (of shifting and unloading, respectively), have been verified in both rigid body simulations (using the combined moments discussed below), as well as in the surface stimulation experiments used to validate the models (discussed below).

[0075] The above discussion suggest that activation of even a moderately atrophic HS muscle group in a seated patient may be sufficient to unload ischial pressure.

[0076] Extensor strength was measured before and after anesthetic block of the sciatic nerve (HS), allowing the isolated effect of gluteus maximus (GM) to be measured. Following sciatic nerve block, hip extensor strength was reduced by approximately 50% across all postures, with the hamstrings accounting for approximately one third of total measured extensor torque (approximately 80Nm at 90°).

[0077] Comparison of extension, external rotation and adduction moments: For maximal GM stimulation in a seated posture, the moments in abduction (9.9Nm) and external rotation (7.1 Nm) can be predicted to be substantial (46% and 33% respectively) as compared to the extensor moment (21.7Nm). HS stimulation may offer a much larger extensor moment (97.3Nm) with only a small degree of internal rotation (2.6Nm). The predicted adduction component may be moderate (52.2Nm) but is of less concern than abduction because it would be stopped in the midline by the contralateral limb or by employing a commercially available knee spacer where necessary. Also, it should be noted that clinically the desired weight shifts may be more moderate than the extreme scenario analyzed here. The full pattern of these torques obtained in all three axes for each of the muscle groups was applied to the musculoskeletal mechanics model. These rigid body simulations support that HS activation would result in ischial elevation, while GM activation could not and would instead only cause lateral sliding of the thigh, which could tend to increase shear.

[0078] SURFACE STIMULATION VALIDATION

[0079] A. Methods

[0080] The rigid body analysis of musculoskeletal mechanics described above may suggest that the extensor torque produced by HS stimulation can relieve ischial pressure ipsilaterally and perhaps even contralateral^ to some degree. This is contrary to the traditional focus on NMES of GM stimulation. To confirm the predictions of this model empirically, we assessed the mechanical effects of these muscles on seated posture and pressure distributions by analyzing surface stimulation data from ten subjects collected by our group previously. We then validated our findingswith a more recent experiment on an additional one subject by syrface stimulation of GM, HS and quadriceps (Quads).

[0081] The first set of data was obtained previously using surface electrodes to stimulate hip extensors individually and in various combinations in ten healthy subjects. Using a seat cushion with an 8 x 8 grid of air-filled cells, seated interface pressures were measured both at rest, and during stimulation of different muscle configurations and voluntary shifting. Surface NMES used a symmetrical biphasic waveform with a frequency of 35 pps and a 250 μs phase duration. Tests were randomized, and each was run three times to determine average rest and stimulation values. The muscle combinations assessed were: 1 ) Quads + GM; 2) Quads + gluteus medius; 3) HS + GM; 4) GM; 5) Quads; and 6) voluntary weight-shifting.

[0082] We noted that HS + GM stimulation was statistically no different from voluntary weight-shifting. These data have been reanalyzed here and compared and combined with the surface stimulation results collected subsequently as follow. Surface stimulation of the hip extensors was performed in a single healthy subject (male, 39 y/o, 76 Kg, 5' 8"). Oval electrodes (1.5" x 2.5" Gentle Stim R Plus; Medical Devices Intl, Saint Louis, MO) were placed over each muscle's bulk with long axes perpendicular to the fibers, 1.5" apart. The subject was seated in a wheelchair with foot and arm rests: thighs were flat; hips, knees, ankles and elbows were flexed to 90° and calves were restrained. Interface pressures were measured using an 18" x 18" (36 x 36 cell) array of sensors (X36 System; XSensor Technology, Calgary, AB, Canada) placed between the subject and a standardized 4" high-density foam cushion (45 kg/ms), on top of a flat hard board. This system uses a thin vinyl interface mat containing capacitive elastomeric sensors. Each of the 1 ,296 sensing elements = 0.5 in² dielectric between two conductive elements. Calibrated accuracy is ±1.3 kPa (10 mmHg) for both observed pressure measurements and inter-trial comparisons. Mitigation against hysteresis (due to compression and subsequent decompression) through software calibrations Range used = 1.3-26.7 kPa (10-200 mmHg). Data was collected in real-time at 10 Hz per cell (13 kHz for the full mat). Stimulation used a symmetrical square waveform at 35 pps delivered as 10 s trains interspersed with 10 s pauses (FastStart EMS; Vision Quest, Irvine, CA). Individual pulse duration was fixed at 300 μs and amplitude was varied between 60-100 mA to control recruitment. Seating pressure distributions were recorded to obtain average records from 3 runs each of GM, HS and Quads stimulation, individually and in all combinations.

[0083] The data was reanalyzed and compared and combined with the new surface stimulation results. Two indices of seating pressure were defined and extracted from the data using Matlab™: 1 ) MPZ = Mean Pressure Zone: user-defined areas in each quadrant for the comparison of average pressure at rest and during stimulation. They were the same size and shape in all four quadrants. Their sizes were normalized at 80% of the contact surface beneath the ipsilateral ischium (i.e. on the side of stimulation). For the buttock quadrants they were centered under the ischiae, and for the thigh quadrants they were placed symmetrically at the most distal edges of measured thigh contact. The average pressure in these zones was compared before and after stimulation. 2) PPA = Peak Pressure Area: an area of those cells with pressures exceeding 8kPA (60mmHg) or higher, beneath the ipsilateral buttock. This may be the maximal surface pressure for which we may predict capillary flow can be maintained in the deep soft tissues. The PPA's size (area) and position (% overlap and centroid shift) were compared at rest and during stimulation. PPA centroids were calculated using a weighted average by area as follows:

X = - and Y = ^-L-

A_τ Λ_τ

where X and Y = coordinates of the centroid, x. and y. = center of mass for each PPA cell (= center of area for this 2D system), a. = area for each cell (0.25 in₂), product %a, = first moment of area relative to y axis, product ya, = first moment of area relative to xaxis, and AT= total PPA area. [0084] Because ai is constant for each cell and AT = aix n (where n = total number of PPA cells), these formulae reduce to:

X =

[0085] B. Results

[0086] Comparing the PPAs in the three examples (areas where pressures exceed the 60mmHg at which capillary occlusion may likely occur in the deep tissues): HS stimulation demonstrates an 88% reduction in PPA size and in PPA overlap, with only one cell remaining unrelieved through both rest and stimulation phases(i.e. above 8kPA (60mmHg)). The middle and R. panels demonstrate progressively worsening results, with the PPA size and % overlap progressively increasing, and the centroid shift progressively decreasing.

[0087] Analysis of the combined MPZ data in this way demonstrated that SE stimulation produced a significant reduction in seated interface pressures as our models predicted (-26% ipsilaterally), together with a mean reduction in total contact area of (mean -25%).

[0088] Surprisingly, GM stimulation actually increased the recorded seating pressures ipsilaterally (mean +16%), with a small reduction in contact surface area (mean -9%). The distribution of pressures suggests that the bulging of the stimulated GM muscle served to concentrate seating pressure in a smaller region ipsilaterally, even slightly off-loading the contralateral buttock (-4%). Conversely, HS stimulation reduced seating pressures both ipsilaterally (mean -26% as above) and contralateral^ (mean -8%), while increasing pressures under the distal thigh. [0089] Regardless of what torques are generated at the hip, the weight of the torso may remain unchanged; only its distribution over different portions of the seating surface can change. Such weight shifts to the thighs may therefore indicative of successful unloading at the buttocks. They may be of little concern because these areas have no boney prominences and may not be at risk for PUs. They may be the normal areas to which mobile individuals can unload when shifting weight.

[0090] Combined PPA data were compared at rest and during stimulation in terms of size and position: PPA size (area): HS stimulation resulted in almost complete elimination of PPA area (mean -94%; from mean 4.0 in∑ at rest to 0.25 in∑ during stimulation); while GM stimulation resulted in over a trebling of PPA area (mean +213%; from mean 4 in∑ at rest to 12.5 in∑ during stimulation). PPA position (% overlap and centroid shift): HS stimulation resulted in only a 6% PPA area remaining within the at-rest PPA area; while with GM stimulation 100% of the PPA area overlapped with the at-rest PPA area (i.e. none of the at-rest PPA area was relieved during stimulation). The centroid shifts with HS and GM stimulation were both similar (mean 0.5" and 0.45" respectively), but due to a large PPA area with GM stimulation (12.5 irte) as opposed to a much smaller one with HS stimulation (0.25 iru?), the significance of centroid shift as a predictive index is unknown at this time. Larger data sets will help clarify its predictive value.

[0091] Anecdotally, with the hip flexed at approximately 90°the greatest effect of GM stimulation alone (at a 3+ clinical muscle strength) was modest external rotation and abduction of the thigh rather than strong hip extension, in keeping with our model predictions.

[0092] The results of both the simulations and the surface stimulation experiments suggest that although GM is the dominant hip extensor during walking, even tetanic activation of a healthy GM muscle may unlikely to provide sufficient hip extension to elevate the ischium and reduce the seated pressures sufficiently to avoid ischemic damage to soft tissues. When seated, GM stimulation was found to provide only -20 Nm or about 33% of the -60 Nm required for unloading, whereas HS stimulation provided -100 Nm or about 170% of the required torque. These findings are somewhat in keeping with those derived experimentally by Waters et al., who measured the relative strength of the HS as extensors of the hip in healthy subjects (n=8). Extensor strength was measured both before and after anesthetic blockade of the sciatic nerve (to HS), allowing the combined effects of the hip extensors to be compared with and without the HS contribution, respectively. At 90^° of hip flexion the HS contribution to extensor torque was mean -80 Nm, although the interpretation is complicated by difficulties experienced obtaining complete blocks.

[0093] The net effect of GM's modest extensor moment and bulging of the muscle between the ischium and the seat may increase the pressure, although the effects in SCI patients might be somewhat less as a result of disuse atrophy. Further, relatively substantial GM torques in abduction and external rotation may produce lateral shifts in seated posture, resulting in shear forces that could exacerbate pressure damage (Fig. 4). The net effect of HS stimulation, however, should offer substantial pressure reductions under the ischium with relatively little lateral motion of the thigh (Fig. 4). PU risk has long been recognized as proportional to the product of pressure intensity and duration. Soft tissues should therefore be able to handle even higher pressure peaks when relieved by intermittent periods of low pressure during which circulation is reestablished. The question then arises as to what is actually accounting for the favorable results reported clinically with GM stimulation. It seems more plausible to attribute them to increased muscle thickness, vascularity, and contact surface area, rather than weight shifting itself.

[0094] GM stimulation may likely be of value for reducing such atrophy and improving vascular capacity. The latter mechanism may account for the favorable results reported clinically for GM stimulation. Substantial pressure reductions may be achieved by SE stimulation under both ischiae. Therefore, two BIONs may be implanted in each patient: one adjacent to the proximal sciatic nerve (to achieve BAS via SE stimulation), and the other adjacent to the inferior gluteal nerve (to achieve improved tissue health and vascularity via GM stimulation). Both nerves are easily and simultaneously accessible during gluteal rotation flap surgery. Supplemental clinical implications of this approach are discussed in the companion appendices I and II.

[0095] The patients at risk for PUs may require bilateral stimulation of one or both muscle groups. Because the weight of the trunk is constant, relief of pressure on one part of the seating surface may be accompanied by increases in pressure elsewhere. We were concerned that elevation of one ischium can be at the expense of increased pressure contralateral^, leading to an increased risk of contralateral PUs. Contractions of SE muscles in one leg actually reduced pressure under both ischiae by transferring pressure to the ipsilateral distal thigh. Lower contralateral reduction of seating pressure can be sufficient to provide any useful protection. Hypoxic tissue damage may likely to arise from constant hydrostatic pressure in excess of that required to occlude circulation. Soft tissues may be able to handle even higher pressure peaks that are relieved by intermittent periods of low pressure during which circulation is reestablished. [0096] GM may be the dominant hip extensor when the hip is in an extended posture, such as during upright locomotion and when lying in bed. The potential exists for GM stimulation to extend and abduct the hip and so roll bed-ridden patients axially. This may relieve both ischial and sacral pressures in supine patients and so may be a beneficial treatment for this group of PU patients, too.

[0097] While we have demonstrated detrimental increases in pressures associated with GM stimulation while seated, we may consider also whether there may be any benefits which might outweigh this detrimental effect.

[0098] It is possible that the dynamic pressure redistributions and centroid shifts during GM stimulation could produce benefits that would offset the increase in pressure produced by muscle bulging. Examining 6 pressure interface mappings of seated GM stimulation demonstrated only minimal decreases in contact surface area during stimulation (Approximate mean = -9%), and PPAs that both increased in size and hardly shifted their overlapping area (<20%).

[0099] We considered also whether the dynamic contractile action of gluteus maximus itself could have a benefit to its own general tissue health through a blood pump type effect. If significant amounts of blood were being pumped through the muscle, then this beneficial effect might outweigh the increased seated pressures predicted during the static contraction.

[00100] This may require a useful volume of blood to flow through the capillary bed on the leading wave of the contraction, before the muscle's hydrostatic pressure exceeded the closing pressure for the capillary bed (~30mmHg). Excitation may propagate along muscle fibers according to the conduction velocity of action potentials along the sarcolemma, which is 3-5m/s; activation and contraction of myofilaments may be tied closely to excitation, so the pressure wave may move at a similar velocity.

[00101] We considered two potential beneficial mechanisms: dynamic pressure redistributions and muscle blood pumping.

[00102] 1) Dynamic Pressure Redistributions: Averages of 6 pressure interface mappings of seated GM stimulation demonstrated only minimal decreases in contact surface area during stimulation (mean = -9%), and PPAs that both increased in size and never shifted their overlapping area. We consider these potential benefits disproportionately small relative to the associated very high MPZ pressure increases (+16%). Therefore to include GM stimulation together with HS stimulation for weight shifting seems likely to compromise the large benefits over large areas that HS stimulation alone would offer.

[00103] 2) Muscle Blood Pumping: The dynamic contractile action of GM itself could have a benefit to its own general tissue health. If significant amounts of blood were being pumped through the muscle, then this beneficial effect might outweigh the increased seated pressures predicted during static contraction. This would require a useful volume of blood to flow through the capillary bed on the leading wave of the contraction, before the muscle's hydrostatic pressure exceeded the closing pressure for the capillary bed = ~4 kPa (30 mmHg). This value has been selected based on Landis' classic measurements in finger arterioles (4.3 kPa (32 mmHg)) ₃₉ , and we recognize that closing pressures for capillaries in gluteal muscles will probably vary from this, as well as across individuals, depending on their age, health status, etc. However, the order of magnitude is acceptable for our purposes here (even allowing for variability of 50%+). Excitation propagates along muscle fibers according to the conduction velocity of action potentials along the sarcolemma (3-5 m/s); activation and contraction of myofilaments is tied closely to excitation, so the pressure wave should move at a similar velocity. Erythrocyte capillary transit velocity at rest = 1 mm/s; and after exercise increases to 4 mm/s. Arteriolar pressure that drives this flow rate is -6.7 kPa (50 mmHg).

[00104] Assuming we can achieve tetany with NMES (for the most rapid blood pumping effect), the maximal force from tetanically contracting skeletal muscle is -30 N/crri2 = 300 kPa (2,250 mmHg). Assuming the flow rate is proportional to this pressure head (and would not be limited by erythrolysis, intimal tears, turbulence, etc.), the 4 mm/s capillary flow rate from a 6.7 kPa (50 mmHg) head of pressure described above can be extrapolated to a capillary flow rate of 180 mm/s. The pressure wave traveling at 3 m/s would overtake and occlude the blood flow advancing at 180 mm/s within 60 ms, at which time there could be no further blood flow / pumping (because the pressures equilibrate in the contracted portions of the muscle at this time). Furthermore, this analysis assumes that tetanic contractile force would be achieved instantly, when in reality there is a rise-time of 50-100 ms, during which there would be even lower pumping pressure available.

[00105] Muscle contractions have been used to assist in pumping blood out of the capillary bed and veins to reduce stasis, but this works only if the hydrostatic force between contractions is sufficiently low to allow arterial pressure to refill these vessels.

[00106] The analysis above bears on the interpretation of recent experiments to determine the effectiveness of NMES for preventing DTI. Solis & Mushahwar, et al. elicited DTI in rats by applying constant loads of 38% of their body weight (expected unilateral loading in seated individuals) to the quadriceps, for 2 hours, with a 3 mm diameter indenter. Experimental groups received intermittent NMES via nerve-cuff electrodes during this constant pressure application. In vivo assessment of deep tissue health was performed using MRI (for detecting muscle edema and oxygenation), 24 hrs following pressure application.

[00107] The authors concluded that intermittent NMES significantly reduces the amount of DTI by increasing the oxygen available to the tissue and by modifying the pressure profiles of the loaded muscles; and proposed that this occurs via two mechanisms: 1 ) periodic restoration of blood flow, reducing injury due to long periods of ischemia and subsequent reperfusion; and 2) reshaping of the underlying muscle, thereby reducing the high stress levels experienced at the muscle-bone interface. However, because the pressure was only exerted over a very small area (0.07 crru?), we believe it likely that stimulation was in fact relieving the muscle by simply intermittently removing it from the pressure zone completely (which is not feasible for GM stimulation in seated individuals). In that case, the muscles would not be subjected to constant pressure as the authors desired, but instead it would be the adjacent unstressed muscle that would briefly be taking the increased stresses that the authors recorded during stimulation.

[00108] Finally, in a single human volunteer, this group measured the changes in GM tissue oxygenation and in surface interface pressures that resulted during GM surface stimulation. Because of space limitations within the MRI scanner, muscle compression during "sitting" was simulated by adding a mass (30% of body weight) over the pelvis of the subject who was lying supine within the scanner with hips extended rather than flexed. A -4% reduction in tissue oxygenation (from baseline) was noted during compression of the buttocks, with a -6% increase (from baseline) after GM activation. It was reported that the surface pressure profiles of the loaded muscles were redistributed and the high-pressure points (over the sacrum) were reduced during surface NMES.

[00109] This is consistent with our analysis of extensor moments for this supine posture. These findings support a mechanism for prevention of PUs in bed-ridden patients that we have proposed previously. As GM is the dominant hip extensor when the hip is extended, such as during upright locomotion and when lying in bed, the potential exists for GM stimulation to extend and abduct the hip and so roll bedridden patients axially. This may relieve both ischial and sacral pressures when supine and so may be a beneficial treatment for this other important group of PU patients.

[00110] Erythrocyte capillary transit velocity at rest = 1 mm/s; and after exercise increases to 4mm/s. Arteriolar pressure that drives this flow rate may be ~50mmHg. Assuming we can achieve tetany with NMES (for the most rapid blood pumping effect), the maximal force for tetanically contracting skeletal muscle may be ~30N/cm2 = 30OkPa = 2,250mmHg. Assuming the flow rate is proportional to this pressure head (and would not be limited by erythrolysis, intimal tears, turbulence, etc.), the 4 mm/s capillary flow rate from a 50mmHg head of pressure described above can be extrapolated to a capillary flow rate of 180mm/s. The pressure wave traveling at 3m/s may overtake and occlude the blood flow advancing at 180 mm/s within 60ms, at which time there can be no further blood flow / pumping (because the pressures equilibrate in the contracted portions of the muscle at this time).

[00111] Furthermore, this analysis assumes that tetanic contractile force may be achieved instantly, when in reality there may be a rise-time of 50-100ms, during which there may be even lower pumping pressure available. Muscle contractions may have been used to assist in pumping blood out of the capillary bed and veins to reduce stasis, but this may work only if the hydrostatic force between contractions is sufficiently low to allow arterial pressure to refill these vessels.

[00112] The NMES have been considered for gluteal stimulation in attempting to achieve weight shifting. It has been determined that Gluteal stimulation is reserved for muscle conditioning, during non-weight bearing periods, and that hamstring stimulation alone may be used for pressure relief during periods of weight bearing.

[00113] Gluteal stimulation could be valuable for muscle conditioning (by reducing disuse atrophy and improving circulation), but bulging of the active gluteus maximus appears to aggravate surface interface pressures without providing sufficient hip extension to elevate the ischium. Thus at least two independently controllable stimulation channels may need to be provided to address ipsilateral PUs; four may be necessary to protect both sides.

[00114] Ultimately it may be desirable to use NMES to prevent both disuse atrophy and the formation of PUs in the first place. Such prophylactic use may be facilitated by nonsurgical injection of BIONs, their usual mode of implantation, in patients who do not already need open surgical repair at the implantation site. The selective activation of the target muscles is also within the scope of the present invention.

[00115] As discussed above while neuromuscular electrical stimulation (NMES) of gluteus maximus may likely to reduce disuse atrophy and improve circulation, it may be neither required nor desired to achieve hip extension through NMES, in order to prevent PUs. Instead hamstring stimulation may be necessary and it could be sufficient to provide hip extension to relieve hydrostatic pressures in the soft tissues under the ischium. [00116] We now compliment our surface stimulation findings above with the finite element modeling of the deep buttock soft tissues in order to identify the promising sites and strategies for such treatment.

[00117] The surface effects of NMES, measured in real-time, were used to drive somatotype-appropriate finite element models of the deep tissue layers in the buttock, enabling the prediction of internal tissue pressure changes in response to recorded surface loading pressures. Average surface pressures of < 60mmHg could be associated with sufficient unloading in all somatotypes to reestablish intramuscular blood flow. The combination of the surface stimulation experiments with finite element modeling suggests a method of treatment employing chronically implanted, wireless microstimulators (BIONs™) to substantially reduce occurrences and recurrences of PUs.

[00118] Soft Tissue Biomechanics

[00119] Now that we have covered the clinical problem of pressure ulcers (PUs) and an analysis of the relevant musculoskeletal biomechanics, we may describe identification of the desired targets and stimulation parameters for relieving internal soft tissue pressures in seated paralyzed patients by neuromuscular electrical stimulation (NMES).

[00120] It has been demonstrated that while gluteus maximus (GM) stimulation may likely to reduce disuse atrophy and improve circulation, it may be neither required nor desired to achieve hip extension in seated subjects. In order to achieve substantial unloading of pressure on the skin under the ischium of a seated subject, it may be necessary to activate the hip extensor hamstring muscles innervated by the sciatic nerve (sciatic extensor (SE) stimulation). [00121] In an exemplary embodiment measurements of skin-surface interface pressures were combined with finite element modeling of the biomechanics of the buttock soft tissues in order to identify promising sites and strategies for NMES. The surface effects of NMES, measured in real-time, were used to drive somatotype- appropriate finite element models of the deep tissue layers in the buttock, enabling us to predict internal soft tissue pressures in response to recorded surface loading pressures. This indicates that average surface pressures of approximately < 60mmHg may be required during unloading, in all somatotypes, to adequately facilitate intra-muscular blood flow. The combination of surface stimulation experiments and finite element modeling has been used to design a detailed research protocol for the prevention of recurrences of PUs in spinal cord in injured (SCI) patients via chronically implanted, neuromuscular microstimulators (BIONs™).

[00122] Method

[00123] Pressures at the seating surface may be easily measured (As also described in surface stimulation experiments in the companion appendix), but determining their consequent effects in the deep tissue layers may be non-trivial. Tissue pressure sensors can be mounted on surgical drains but they may interfere with actual tissue pressure distributions. They also may have to be removed well before patients are able to resume weight-bearing on the surgical site. Instead, noninvasive measurements obtained from the surface stimulation experiments were used to develop, drive and validate finite element analysis (FEA) models of the buttock soft-tissues under different loading conditions, and for different somatotypes.

[00124] These models may predict the internal pressures in the different tissue layers that may accompany the measured surface pressures, both as a guide for setting treatment parameters and to offer insights into what may occur in the deep tissues. Two finite element models were developed: 1 ) an axisymmetric 2D hemi- buttock model in MSC.AFEATM, and 2) a 3D buttock model in ANSYSTM. The 2D axisymmetric model may be preferable due to lower computational demands and higher accuracy than the 3D model (as is common with FEA techniques), and was validated using the surface stimulation experiments referred to above, by comparisons of simulations with clinical patterns and with theoretically calculated results, as well as with pressurized meat and graded seating experiments.

[00125] Internal tissue pressure changes may be predicted in a non-invasive way, in real-time, and whenever needed. An embodiment may compute what reduction in measurable surface pressures could avoid circulatory occlusion in the muscles.

[00126] Intramuscular pressures may fall below the arterial capillary closing pressure of -30 mmHg. Systems such as the skin, fat and muscles have a combination of properties that make them difficult to model using FEA. In particular, these fluid-filled tissues may be incompressible but may have large and nonlinear compliances. Their mesh dimensions may undergo large deformations from unloaded to maximally loaded conditions, which may result in compounding of errors and computational instability. They can also be difficult to characterize due to model friction, shear, and dynamic viscous creep etc. between discrete layers with different properties.

[00127] We studied the behavior of a static model over the range of loads for which it was computationally stable. These were the lower pressure portions of incremental loading curves. The higher pressure regions, where the model could become unstable, are associated with vessel occlusion and so may not be of relevance here. The changes in internal tissue pressures under these conditions have been extrapolated to those expected under normal seating pressure. [00128] An axi symmetric model was created in MSC.AFEATM (MSC. Software Corp), which may consist of MSC. Marc (the FEA solver) and MSC.Patran (the FEA pre- and postprocessor). The buttocks were treated as axially symmetric structures about the ischium. The centrally located bone is covered by muscle, then fat and finally skin outermost. Contacts between layers were defined as follows: bone-tissue = glue; tissue-tissue = glue; seat-skin = stick-slip friction (using nodal stresses). Analyses were run as non-linear, static, axisymmetric to 3D, solving for large displacements and large strains, and using the full Newton-Raphson iteration method.

[00129] In calculus this method is often referred to as Newton's method for extracting a root of a polynomial. It was used here to generate load-displacement curves for non-linear problems where the initial shape may be unknown; and involves history dependent responses, so that the solution may be obtained as a series of increments, with iterations within each increment to obtain equilibrium. Material properties used are detailed in Table 5.

Table 5 - Material properties and Behavior.

[00130] Because the analysis was limited to the low-deformation range, the materials were modeled as elastic and isotropic. We decided to limit the mesh to the 500 element limit of the educational version of this commercial software.

[00131] This model was validated in 4 ways: 1 ) Repeated simulations under different configurations demonstrated that the model behaved intuitively as one may expect from clinical pathology. The loaded system demonstrated two high pressure zones: one in the skin and one in muscle (close to hard bone and associated with its relatively small surface area). This pattern corresponds to the frequent clinical finding of PUs beginning deep in the muscle, and then suddenly breaking down through the skin. 2) Model data during simulations compared favorably with theoretical calculated results: within the range of interest (low deformations), the model's force, displacement and contact surface area values fell within 10% of calculated values. 3) Model data during simulations compared favorably with experimental data obtained during pressurized meat experiment runs (at low deformations). This involved measuring the hydrostatic pressures in fresh porcine muscle under known compressive loads in vitro. 4) Finally, model data during simulations also compared favorably with experimental data during the low deformation portions of a graded seating experiment. This involved lowering a subject from a gantry onto a pressure sensor mat and recording contact areas and pressures at progressively increasing degrees of seating contact and ischial loading.

[00132] Implementing the model simulations involves 3 steps: 1 ) sequential unloading until a goal average intramuscular pressure (Pm) of < 30mmHg is reached; 2) discretizing the solution at the surface; and 3) determining the average surface pressures (Ps) that are expected to correspond to these intramuscular pressures (Pm). Because the FE model is unitless, this process can be customized to each patient's overall dimensions (by scaling) and somatotype (by reassigning element material properties to alter tissue layer proportions). This then may provide a look-up table, specific to each patient, by which their intramuscular pressures (Pm) can be estimated in real-time from recorded surface pressures (Ps) as described above.

[00133] Simulations were carried out for Sheldon's 3 different somatotypes, using estimated skin:fat:muscle ratios of 1 :7:3 for endomorphs, 1 :3:7 for mesomorphs and 1 :1.3:1.3 for ectomorphs. For each of these example somatotypes ischial loading was compared with contact surface area, and intramuscular pressure changes (Pm) were compared to surface pressure changes (Ps) during progressive loading. Average surface pressures (Ps) at which intramuscular capillaries would be open (Pm < 30mmHg) were determined for each group.

[00134] Results

[00135] Figs. 1A-B provides incremental loading curves for models of different somatotypes. In panel of Fig. 1 B, comparing ischial loading with contact surface area, an expected non-linear plateau pattern is noted at high ischial loads for all 3 somatotypes (upper panel). For the low-loading conditions that are of interest here

(lower panel), note that while endomorphs and mesomorphs have very similar contact curves, ectomorphs have the smallest contact surface areas for the same ischial loads. This is in keeping with the higher surface pressures (Ps) that are seen in Fig. 1 B in this group, as well as with the higher risk of PUs found clinically in this group.

[00136] In Fig. 1 B the changes in intramuscular pressures (Pm, solid curves) and surface pressures (Ps, dashed curves) are compared with progressive loading for the 3 somatotypes. While all 3 groups demonstrate similar average intramuscular pressures, note again that ectomorphs differ clearly from endomorphs and mesomorphs (upper panel). In the lower panel at low-loading conditions, for intramuscular pressures to fall below 30mmHg (capillary closing pressure), surface pressures fall below 60 mmHg in endomorphs and mesomorphs; but below 100 mmHg in ectomorphs.

[00137] Different Somatotypes

[00138] Ectomorphs are thin and bony and generally may be at higher risk for PUs than endomorphs and mesomorphs. Counter-intuitively, the model predicts that intramuscular capillary flow will be maintained at a higher surface pressure (<100mmHg) in ectomorphs than in meso- and endomorphs (<60mmHg); the difference just about compensates for the different surface pressures predicted for these body types. This suggests that ectomorphs may be at greater risk for superficial PUs specifically (associated with friction, shear and cutaneous pressure increases), but at similar risk for deep PUs. Superficial PUs are easier to detect and are associated with a lower morbidity than deep ones. The greater muscle bulk in endo- and mesomorphs can provide additional protection of the skin through a larger contribution of musculocutaneous perforating vessels. The increased strength of their muscles might make NMES more effective, but the effect could be more than counteracted by their increased trunk mass that may be elevated. It would appear, however, that setting a measurable goal of Ps < 60mmHg in all patients could provide protection for all patients.

[00139] Implications for NMES Treatment Protocols

[00140] The preclinical work described above has been directed toward planning the BION implant locations, to anticipating their clinical effects, and to developing objective criteria for setting adequate stimulation parameters. In addition to using tetanic SE stimulation to achieve relief of surface pressure, effective treatment may likely also require a period of conditioning of atrophic muscles. Both the SE and GM muscles may likely be severely atrophic in patients with long-standing SCI.

[00141] Disuse atrophy of SE muscles may well cause them to have insufficient strength and fatigue resistance to provide sufficient, repetitive unloading of the ischium over the course of the desired period of sitting in a wheelchair. Disuse atrophy of GM muscles reduces the padding and the availability of musculocutaneous blood supply for the overlying skin.

[00142] Furthermore, the GM muscle needs to heal from the surgical disruption and suture repair of the rotation flap. Fortunately, there is evidence suggesting that substantial conditioning is feasible in both of these muscle groups during the postoperative recovery period, without requiring high forces or disruptive movements. Previous studies in a rat model of disuse atrophy have shown that unfused twitch stimulation that produces only low forces is effective in preventing atrophy, probably through activation of the calcium kinase pathway responsible for the maintenance of muscle mass. [00143] Furthermore, clinical studies using BIONs to prevent shoulder subluxation following stroke, and to strengthen quadriceps in patients with knee osteoarthritis, have used 2-5Hz stimulation with success. Low frequency stimulation can be applied to condition both SE and GM muscles during the 5-8 week post-operative period during which the skin and muscles cannot be subjected to large contractile forces or sitting pressures. Proposed initial treatment parameters are provided in Table 6. These may be titrated as required to achieve the desired outcomes (described above).

Table 6 - Proposed Initial Stimulation Parameters

[00144] Additional Considerations for Clinical Therapy

[00145] The sciatic nerve is a mixed sensory and motor nerve that innervates not only the SE, but also most of the distal leg. Although SCI patients are largely insensate, postural shifts, as well as reflex effects on spasticity and autonomic dysfunction may be evaluated clinically to identify safe and effective stimulus parameters for the sciatic site. At this anatomical level, the outermost and most medial nerve fascicles supply the muscles of interest and may have lower thresholds than the deeper fascicles. Careful exploration with a handheld probe stimulator may help to identify a placement of the BION that will provide some degree of selective stimulation. The relatively low frequencies of stimulation that may be required to produce fused muscle contractions are also may unlikely cause large reflex effects from proprioceptive afferents (Table 6). Gradual ramp-up of each stimulation train could reduce mechanically evoked spasticity.

[00146] It remains to be determined whether patients at risk for PUs will require bilateral stimulation of one or both muscle groups. Because the weight of the trunk is constant, relief of pressure on one part of the seating surface might be accompanied by increases in pressure elsewhere.

[00147] As discussed above, we were concerned that elevation of one ischium may be at the expense of increased pressure contralateral^, leading to an increased risk of contralateral PUs. However contractions of SE muscles were predicted and demonstrated to reduce pressure under both ischiae by transferring pressure to the ipsilateral distal thigh. It remains to be seen, however, whether the much lower contralateral reduction of seating pressure could be sufficient to provide any useful protection.

[00148] Hypoxic tissue damage may likely arise from constant hydrostatic pressure in excess of that required to occlude circulation. Soft tissues can handle even higher pressure peaks that are relieved by intermittent periods of low pressure during which circulation is reestablished. One caveat is that the shifting motion may increase shear and abrasion of the skin, which may be known to be contributing factors in superficial PUs. Even with NMES, patients may need the usual monitoring, skin care and well-designed garments and cushions.

[00149] Surface effects of NMES can be measured in real-time using conventional pressure arrays. These data have been used to develop, drive and validate somatotype-appropriate finite element models of the deep-tissue layers in the buttock. Such models are patient-specific and provide predictions of internal tissue pressure changes in response to recorded surface loading pressures in an otherwise complex and pathological biomechanical system. Finite element analysis indicates that achieving average surface pressures of < 60mmHg may be required during unloading to adequately facilitate intramuscular blood flow, and that this holds true for all somatotypes.

[00150] By combining surface stimulation experiments (As also described in the companion appendix I) and finite element modeling for planning treatments, we have developed a new treatment paradigm for pressure ulcer prevention through NMES:

[00151] 1. Use multiple sites of stimulation during non-weightbearing to condition atrophic muscles (e.g. GM & SE);

[00152] 2. Use SE stimulation during weight-bearing as actuators to achieve unloading via hip extension torque;

[00153] 3. Determine stimulation parameters that will adequately relieve internal soft tissue pressures, by aiming for peak seating pressures < 60mmHg and redistributions as (Also discussed in the companion appendix I).

[00154] Musculoskeletal and soft tissue models support the aspect having the BION Active Seating (BAS) will improve pressure distribution and tissue health. The models and analyses discussed above can be used to guide clinical treatment and evaluate results, for the eventual application of BAS to prophylaxis, with percutaneously injected, bilateral implants used to prevent PUs from forming.

[00155] The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use pressure ulcer prevention (PUP) or treatment system and method. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the pressure ulcer prevention (PUP) or treatment systems and methods. Thus, pressure ulcer prevention (PUP) or treatment systems and methods are not intended to be limited to the embodiments shown herein but are to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

CLAIMS:

1 ) A method for preventing pressure ulcer in an immobile patient's body comprising:

a) implanting one or more leadless miniature neuromuscular electrical stimulators near one or more target muscles in the patient's body; and

b) activating remotely the one or more leadless electrical stimulators so as to stimulate the one or more target muscles and shift the patient's weight.

2) The method of claim 1 , further comprising testing the efficacy of the stimulations of the one or more target area of the patient at rest and during stimulation before implanting the one or more electrical stimulators.

3) The method of claim 2, further comprising assessing pressure relief of the one or more target areas of the patient's body based on the stimulations.

4) The method of claim 3, further comprising generating a relief protocol that is tailored to the patient based on the testing.

5) The method of claim 3, further comprising measuring pressure that is achieved by the stimulations at skin-surface of the patient's body. 6) The method of claim 5, further comprising generating data that is related to pressure that is achieved by the stimulations at internal tissue layers of the patient's body.

7) The method of claim 1 , wherein the one or more target areas comprising muscles that are selected from gluteus maximus and/or sciatic extensor muscles.

8) The method of claim 2, further comprising applying a skeletal biomechanics model and/or soft tissue biomechanics model so as to identify an effective target area and/or effective stimulation parameters.

9) A system for preventing pressure ulcer in an immobile patient's body comprising:

a) one or more leadless miniature neuromuscular electrical stimulators configured to be implanted near one or more target muscles in the patient's body and to stimulate the one or more target muscles so as to allow shifting of the patient's weight; and

b) a controller configured to activate the one or more leadless electrical stimulators by remotely activating the implanted one or more electrical stimulators.

10)The system of claim 9, further comprising one or more stimulation systems configured to test the efficacy of the stimulation of the one or more target area muscles at rest and during stimulation before implanting the one or more electrical stimulators, wherein the stimulation systems are selected from a group: surface stimulation system, needle electrode stimulation system and electromagnetic stimulation system.

1 1 )The system of claim 10, further comprising a display device that is configured to display information related to pressure relief at the stimulated one or more target area muscles of the patient's body.

12)The system of claim 9, further comprising a relief protocol that is configured to generate a custom muscle stimulation program for the patient based on the applied stimulations.

13)The system of claim 10, further comprising one or more sensors configured to measure the pressure that is achieved at skin-surface of the patient's body.

14)The method of claim 13, further comprising a computing device configured to generate data that is related to pressure that is achieved at internal tissue layers of the patient's body.

15)The system of claim 9, wherein the one or more electrical stimulators are configured to stimulate gluteus maximus and/or sciatic extensor of the patient's body.

16)The system of claim 10, further comprising one or more skeletal biomechanics models and/or soft tissue biomechanics models, configured to provide data in identifying effective one or more target areas and/or effective stimulation parameters.

17) The system of claim 16, wherein the one or more skeletal biomechanics models further comprising one or more rigid body type analysis models and/or interactive musculoskeletal models.

18)The system of claim 16, wherein the soft tissue biomechanics models comprising finite element analysis models.

19)The system of claim 10, further comprising one or more electronic or mechanical pressure sensor array that is configured to measure surface interface pressure that is achieved at rest and/or during the stimulation.