Journal of Rehabilitation Research and Development

Journal of Rehabilitation Research and Development

Rehabilitation R & D Progress Reports 1986

VII. Wound and Fracture Healing

VII . Wound and Fracture Healing

Electrical Stimulation for Augmentation of Wound Healing

Scott R. Crowgey, M.D., and Steven M. SharpeVeterans Administration Research and Development, Decatur, GA 30033Sponsor: VA Rehabilitation Research and Development Service

Purpose—This project will attempt to identifyaspects of the wound healing process that maybe augmented by the exogenous influence ofelectromagnetic fields. A theoretical analysis ofthe possible effects of electromagnetic fields onwound healing will include analyses of theinteraction of electromagnetic fields with cellu-lar structures and of the deposition of heat indamaged tissue via exogenously applied energyfields . This analysis will then be used as a basisfor developing a plan for future investigationsinto the potential application of electrical stim-ulation for the augmentation of wound healing.

The initial research will involve a review ofthe literature to identify aspects of wound heal-

ing that could be influenced by electrical stimu-lation. Efforts will then be directed toward de-veloping mathematical models of the possibleelectrical interaction of electromagnetic fieldswith cells and cell structures to determine howthese interactions could be optimized to im-prove wound healing. It is anticipated that theliterature will not contain all the informationnecessary to develop these models . Any gaps innecessary information and data will be filled, ifpractical, using tissue phantom modeling mate-rials, blood, and possibly even primitive tissueculture exposed to a variety of known electro-magnetic environments, using easily construct-ed exposure chambers.

Development of a Mathematical Model of Fracture Healing in Long Bones

Gary Beaupre, Ph.D., and Dennis Carter, Ph.D.Rehabilitation Research and Development Center, Veterans Administration Medical Center, Palo Alto, CA 94304Sponsor : VA Rehabilitation Research and Development Service

Purpose—When a long bone is fractured, itsmechanical properties change significantly asthe bone heals. However there is very littlequantitative information on how mechanicalstresses influence the healing process and onhow the mechanical properties of the fracturesite change as the bone regenerates back to itsoriginal shape and structure.

Much attention has been given to the de-velopment of fixation devices for fractured longbones, and many different mathematicalmodels have been used to evaluate the stressstates and deformations caused by the applica-tion of these devices. The mechanical propertiesassumed at the fracture site influence the re-sults of these analyses. The development of amathematical model to quantify how a bone

heals in the presence of different stress statesand how the mechanical properties change ac-cordingly will offer insights into the fracturehealing process for both normal and pathologi-cal healing situations. A better understandingof fracture biomechanics will provide guidelinesin the design of treatment modalities for frac-tures and fracture nonunions.

To date, there have been numerous histo-logical and morphological studies on the heal-ing process of fractured bones. The healingprocess has been categorized into four basicphases: 1) initial reformation of the vascularsystem with formation of soft callus tissue ; 2)formation of cartilaginous callus ; 3) formationof bone tissue; and 4) remodeling of bone tissue.These sequential phases stabilize and progres-

228

229

Rehabilitation R&D Progress Reports 1986

sively stiffen and strengthen the healing bone.Some data are available on the mechanical

properties of the tissues involved, but nomodels exist that can describe or predict thishealing process . Our hypothesis is that thetransition from the initial phases of fracturedbone to its original structure, in particularfrom the cartilaginous stage to the formation ofbone tissue, is influenced by the local state ofstress and strain within the tissues participat-ing in the healing process . As healing proceeds,the material and structural properties of thefracture are changed, resulting in a progressiveincrease in the structural integrity of the bone.

Progress—The finite element method is beingused to create a mathematical model of a frac-tured long bone. The idealized bone is basedupon the midshaft geometry of the human

femur. The appropriate loading and boundaryconditions are obtained from previous studiesusing: 1) load-sensing implants and prostheses;2) in vitro experiments; and 3) static and dy-namic mathematical models of lower limbforces. An iterative mathematical procedurethat triggers the transformation of cartilageinto bone will be used. This procedure willpermit the temporal and geometric pattern ofhealing to be both predicted and followed.

In this study two- and three-dimensionalnonlinear finite element models of an idealizedfractured bone are being developed. Initial frac-ture callus geometry was determined from pre-vious studies. Preliminary studies suggest thatthe local strain energy density is a likely candi-date for the controlling mechanism that trig-gers the transformation from cartilage intobone.

Bioelectricity in Fracture Healing

John F . Connolly, M .D .; Dennis A. Chakkalakal, Ph.D . ; Louis Lippiello, Ph.D.Veterans Administration Medical Center, Omaha, NE 68105, and the University of Nebraska Medical Center, Omaha68131Sponsor : VA Rehabilitation Research and Development Service

Purpose—To investigate the role of bioelectri-city in fracture healing, two research projectsare under way: "Electrical Osteogenesis: Mech-anisms and Causes of Failure," and "ScientificBasis for a New Protocol in External Fixationof Fractures ." The basic knowledge of bioelec-tricity's role derived from these studies will beused to identify electrical signals that will bemost effective in promoting healing in fracturesthat are difficult to heal . In patients with an os-teoporotic condition, external fixation is pre-ferred over internal fixation . However, externalfixation has a tendency to delay fracture heal-ing. Therefore, we are attempting to identifyelectrical signals that may be applied to pre-vent such a delay.

Progress—We are developing objective methodsfor measuring the rate of return of mechanicalrigidity, bone blood flow, and skin-surface elec-trical activity toward normal in fracture heal-ing in dogs so that these can eventually be used

in humans for diagnosis of problem fractures,for choice of the optimum method of treatment,and for prognosis of all fractures.

We used both internally fixed and external-ly fixed long-bone fracture models (radius andtibia) in dogs that heal normally and those thatare delayed to carry out our studies accordingto the following two themes:

A) Characterize these long-bone fracturemodels through biomechanical, physico-chemi-cal, vascular, histomorphometric, and biochemi-cal measurements so that we can understanddetails of the differences between a normallyhealing fracture and a delayed-healing one.

B) Investigate roles of electrical activitythat occur in living systems naturally followinginjury and of externally applied electricity inaltering the various features of these fracturesrevealed from the above study.

Preliminary Results—From the studies undertheme A, we obtained the following results : 1) a

230

Wound and Fracture Healing

method for determining an index of rigidity ofthe fractured bone; 2) a method for measuringthe return of the venous flow in bone towardnormal; 3) a predictable relationship betweenthe patterns (with respect to time) of the rees-tablishment of blood flow and mechanical prop-erties; and 4) the relationships between biome-chanical and biochemical features and also be-tween biomechanical features and calcificationin fractures.

From the second set of studies, theme B,the results were as follows: 1) There are twonaturally occurring "batteries" in dogs—one as-sociated with the epidermis of skin and theother with the endosteum of bone—that becomeactivated in response to an injury involvingboth bone and the surrounding tissues. 2) Skin-surface measurements of electrical voltage andcurrent reveal different patterns associatedwith normal and delayed-healing fractureswithin the first 3 to 4 weeks after the injury . 3)Two different DC signals (1 microampere and 7

microamperes) applied during the first weekafter fracture seem to stimulate cell prolifera-tive activity that leads to a significant accelera-tion of bone formation at the fracture site at 7weeks after injury ; two different cell popula-tions or two different mechanisms may be in-volved in this response. 4) Comparison of fivedifferent electrical signals similar to those usedclinically in fracture healing, as applied in longbones in dogs, demonstrate widely different fea-tures (amplitude, frequency, etc .), even thoughthey have comparable clinical efficiency.

Our studies under theme B also suggestthat none of these signals may be the most ef-fective one for fracture healing . We expect thatthe knowledge about the influence of specificelectrical parameters on the various features offracture healing gained from our studies will bea useful contribution to the development ofmore reliable procedures for electrical therapyand prognosis in fracture healing.

Stimulation of Repair of Cortical Bone Transplants by Implantation of PiezoelectricMaterials : A Development Study

G . V. B. Cochran, M .D., M.Sc .D.Orthopaedic Engineering and Research Center, Helen Hayes Hospital, W . Haverstraw, NY 10993, and Surgical ResearchService, Castle Point Veterans Administration Medical Center, New York 12511Sponsor : VA Rehabilitation Research and Development Service ; New York State Department of Health

Purpose—Stimulation of bone healing by mi-croampere electric currents now is a recognizedform of clinical treatment . For this purpose,various devices are available that aim to im-prove the healing of bone in problem cases, par-ticularly for individuals with delayed nonunion.Typically, the stimulatory current is providedby a battery-powered device that delivers cur-rent to bone via electrodes or by an externalelectromagnetic device that induces currents inbone. Our purpose is to develop a novel ap-proach to electrical stimulation of bone healingthat employs microampere currents generatedduring physiological loading by a piezoelectricmaterial that is incorporated as part of an in-ternal fixation plate . Because piezoelectric ma-terials produce an electrical charge under me-chanical loading, we are thus developing an im-

plant that not only will stabilize bone but alsowill act as a self-generating electrical stimula-tion device that requires no separate powersource.

Progress—To date we have designed and testedseveral versions of the "piezo plate ." Initialtests showed that piezoelectric materials placedon the plate (so as to be in direct contact withbone) were not effective, probably becausecharge density over the surface was too low.Accordingly, we developed a device in whichthe piezoelectric material is sealed within theplate and all the charge developed is collectedand delivered to bone via electrodes . During thepast year we concentrated on electronic andmechanical design considerations necessary tocreate a version that could be used for trials in

231


large animals as a precursor to fabricating aninitial model for clinical use.

Preliminary Results—Our animal tests so farhave concentrated on the electrical output ofprototype plates under functional conditions.These tests demonstrated that even on intactbone (such as in a late stage of fracture heal-ing), the deformation produced by limitedweightbearing is adequate to produce electriccurrents in the range known to stimulate osteo-genesis, after necessary rectification and proc-essing by miniature circuitry . Furthermore, we

showed that an implanted plate can be activat-ed by external, low-power ultrasound to gener-ate currents in an optimal stimulatory range.These developments hold promise for reducingthe incidence of nonunion in problem fracturesand may have potential for enhancing fixationof joint replacement prostheses.

Future Plans—The next phase of our study,"Testing of Design Parameters for a PrototypePiezoelectric Internal Fixation Plate," has beenapproved and involves testing the effects of theplate on bone healing in sheep.

Stress Analysis of Internal Fracture Fixation of Long Bones

Gary Beaupre, Ph.D., and Dennis Carter, Ph.D.Rehabilitation Research and Development Center, Veterans Administration Medical Center, Palo Alto, CA 94304Sponsor : VA Rehabilitation Research and Development Service

Purpose—Internal fixation of fractured longbones using metal plates is a standard proce-dure for treating fractures that would be unsta-ble if treated conservatively (e .g., splinted orcaused) . A possible complication resulting frominternal fixation is refracture of the bone afterthe plate is removed . Some clinical studies haveshown refracture rates to be as high as 20 to 25percent.

Internal fixation is used on the long bonesof both the lower (femur and tibia) and upper(humerus, ulna, and radius) extremities . Al-though exact figures for the number of frac-tures treated with internal fixation are notknown, the popularity of this treatment modeis steadily increasing, particularly for treatingforearm fractures where full range-of-motion inpronation and supination may be jeopardizedwith conservative treatment.

Refracture rates for the radius and ulna of10 to 20 percent can be directly interpreted interms of longer healing times with an associat-ed loss of human productivity and increasedmedical costs for reoperations and prolongedphysician treatment.

Although both vascular damage and me-chanical stress shielding have been implicatedas contributing factors leading to refracture, nostudies have assessed mechanical stress shield-

ing using realistic mathematical models . Allstudies to date have assumed that no motion orslip occurs between the plate and the bone.However, numerous in vitro tests have shownthat slipping between plate and bone occurs toa significant degree . By utilizing a nonlinear,mathematical model that permits slipping be-tween plate and bone, we will be able to deter-mine accurately the extent to which stressshielding is responsible for bone refracture.

We hypothesize that refracture after plateremoval is caused by a combination of vasculardamage and mechanical stress shielding. Theimportance of stress shielding has been overem-phasized in all previous mathematical modelsbecause earlier studies have not incorporatedrealistic interface conditions. By including real-istic interface conditions, we may arrive at asignificantly different assessment of theamount of stress shielding. This in turn mayhave significant impact on future plate designand on current plate application techniques.

Progress—Using a nonlinear finite elementmodel, we are capable of simulating slippingmotion between the internal fixation plate andthe fractured bone . Coulomb-friction contactelements are used to model the interface be-tween the plate and bone and between the un-

232


derside of the screw heads and the plate . Screwtightness, friction coefficients, plate material,and plate-bone conformity will be studied para-metrically.

An idealized plated bone model has beendeveloped and analyzed for five possible in vivoloading conditions . The degree of mechanicalstress shielding is shown to be intimately relat-

ed to how tightly the screws are applied. Loosescrews produce minimal stress shielding; tight-er screws produce more stress shielding. Esti-mates of screw tightness at plate explantationcombined with the results of the present studysuggest that previous predictions of stressshielding may be grossly in error.

Quantifying Fracture Healing by Impulse Transfer Functions

R. E. Jones, M .D .; M. G. Strauss, Ph.D . ; R. W. Bucholz, M.D . ; K. L. Lawrence, Ph.D.Veterans Administration Medical Center, Dallas, TX 75216 ; University of Texas at Arlington, Arlington, TX 76019;University of Texas Health Science Center at Dallas, Dallas TX 75235Sponsor: VA Rehabilitation and Research Development Service

Purpose—The need for a technique to quantifyfracture healing and the role it would play inredefining clinical union, identifying delayedunions, and optimizing rehabilitation programsfor accelerated fracture healing is well recog-nized in the orthopaedic community . Currently,only two tests to assess fracture healing areavailable—the roentgenogram and manual ma-nipulation. Both are qualitative. Over the past4 years, we have sought to develop a techniqueto quantify fracture healing in long bones vianatural frequency analysis.

Progress—The natural frequencies of a struc-ture are the frequencies at which the structurewill vibrate with large amplitudes compared toother frequencies. Previous research has shownhow the natural frequencies of the in vitro tibiachange as the depth of a cross-sectional cut in-creases . The effects of the fracture location,callus size, and calcification process on the nat-ural frequencies in vivo, and the relation ofcallus strength are being studied.

In the clinic, 26 patients with tibia frac-tures were followed. For the 24 normally heal-ing patients, a correlation of R = 0.74 existedbetween the days postinjury and the normal-ized natural frequency . Two patients developeddelayed unions and demonstrated differenthealing curves compared to the normally heal-ing patients.

In order to establish the correlation be-tween the natural frequencies and the strength

of the union, a dog model was established, andexperiments were performed . Twenty-one malemongrel dogs underwent unilateral radial os-teotomies. The forelimb was placed in a paddedplastic spoon splint for 6 weeks postsurgery.Starting at 2 weeks postsurgery and at weeklyintervals, each dog was placed in an elevatedtotal body harness so that its limbs would hangclear of any surface; resonance testing was thenperformed on the radii of the unconscious dog.The impact transfer functions of both radiiwere obtained by impacting an instrumentedhammer at the proximal lateral radius whilean accelerometer was held against the anterola-teral radius . Similar to the human study, thetransfer function was obtained using a Fouriertransform. After the designated healing timefor a dog had been reached, the dog was euth-anized, and both forelimbs were promptly re-moved. A 10-cm section from each radius wasremoved. The fracture site was at the mid-length. Each section was subjected to torsionalfailure.

Preliminary Results—The results of the tor-sional tests from 20 dogs showed an increase intorsional strength with increased time until ap-proximately 100 days postosteotomy, at whichtime the torsional strength of the callus wastwice the torsional strength of the intactradius. With the data from only 4 dogs ana-lyzed, a trend appeared. As time progressed,the fracture radius natural frequency squared

233


divided by the normal radius natural frequencysquared, also increased . The linear regressionlines for the torsional strength versus time andthat of the natural frequency versus time arevery similar in slope. This implies that it wouldbe possible to follow the course of healing inlong bone fractures noninvasively by this tech-

nique and to identify abnormal deviations fromthe normal healing pattern sooner than presenttechniques allow . Clinical union may be quanti-tatively assessed in vivo. This method couldalso be used to determine what affects the rateof healing and to tailor an optimal rehabilita-tion scheme to accelerate fracture healing.

Altered Collagen and Wound Metabolism in Non-Healing Diabetic Ulcers

Roger E. Pecoraro, M .D .; John Olerud, M .D . ; Mary Ann Riederer-Henderson, Ph.D . ; Ernest Burgess, M .D . ; Frederick A.Matsen III, M.D . ; Craig Wyss, Ph .D.Seattle Veterans Administration Medical Center, Seattle, WA 98108Sponsor: VA Rehabilitation Research and Development Service

Purpose—This project was designed to test thehypothesis that potentially correctable metabol-ic abnormalities may interact with ischemia,neuropathy, and infection to obstruct healing ofdiabetic ulcers . Subjects include diabetic andnondiabetic patients admitted to the SeattleVAMC Amputation Service who may requirelower extremity amputation as a result of dia-betes and/or vascular disease. We will testwhether recent poor glycemic control, abnormalascorbic acid metabolism, altered zinc availabil-ity to injured tissue, and increased nonenzyma-tic glycosylation of dermal collagen are associ-ated with and potentially responsible for failureof wound healing that leads to amputation indiabetic individuals.

The objective of these cumulative investiga-tions is to identify metabolic abnormalities thatare potentially correctable and that may con-tribute to limb loss and wound failure in dia-betic individuals.

Progress—Glycemic control in diabetic patientsis being documented by measurements of fast-ing plasma glucose and glycosylated hemoglo-bin. Ascorbic acid levels are determined in sam-ples of plasma from all patients and in leuko-cytes and dermal tissue from selected patients,measured by high-performance liquid chroma-tography. Zinc levels are being measured insamples of plasma, leukocyte, and wound tissueby atomic absorption spectrophotometry . Colla-gen fractions are extracted from skin andwound tissue specimens from amputated limbs

for subsequent measurement of the extent ofglycosylation of collagen . Nutritional status isbeing evaluated by a laboratory panel of nutri-tional indicators. Vascular status of diabeticand nondiabetic amputation subjects is beingdocumented by standardized measurements oflimb transcutaneous oxygen tension (TcPO2 )and segmental doppler blood pressures.

To date, we have studied 79 diabetic indi-viduals who have been candidates for or whohave actually received limb amputations. Atotal of 49 nondiabetic amputees, all with pe-ripheral vascular disease, were enrolled . In ad-dition, many of the biochemical and vascularmeasurements were standardized in 10 healthynonsmoking elderly males without diabetes orvascular impairment . Vascular and plasmametabolic measurements have been made in150 control diabetic individuals admitted to thesame hospital but who have not had amputa-tions or lower extremity ulcers.

Preliminary Results—Both diabetic and non-diabetic amputation patients have shown signif-icant deficiencies in plasma zinc and ascorbicacid levels. Preliminary analysis of tissue ex-tracts suggests that ascorbic acid is concentrat-ed in dermal tissues at sites of cutaneous ulcer-ation. Further studies are in progress to deter-mine if tissue concentrations of ascorbate andzinc in these patients are suboptimal for ade-quate wound healing. Experimental iatrogenicmicrowounds have been inflicted on the limbsof a volunteer subgroup of patients 7 days

234


before amputation; histologic evaluations of patients with diabetes, unlike nondiabetics re-those tissues, in progress, will provide a semi-

quiring amputation resulting from advancedquantitative independent index of cutaneous

peripheral arterial disease, often fail to healwound healing to correlate with the metabolic limb ulcers despite adequate cutaneous oxygenparameters .

diffusion.We confirmed the observation that many

Morphological and Clinical Studies of Microwounds in Ischemic Human Tissue

John Olerud, M .D.School of Medicine, University of Washington, Seattle, WA 98195Sponsor : VA Rehabilitation Research and Development Service

Purpose—The purpose of this project is tostudy in a systematic fashion morphologicaland clinical features of small standardizedwounds created on the lower extremities ofnormal elderly subjects and patients withsevere peripheral vascular disease (PVD) neces-sitating amputation.

Progress—We developed a standard humanwound model for evaluating 12 morphologicalevents of repair . The events include presence ofscab, epidermal migration, epidermal closure,epidermal hyperplasia, amount of stratum cor-neum, stratum granulosum, fibrin neutrophiles,monocytes, fibroblasts, collagen, and capillaries .

Wounds from 18 diabetics, 13 patients withPVD, and 11 normal subjects were fully proc-essed and evaluated . The results were enteredon a computer database, and preliminary dataanalysis was done . Additionally, we developed atimetable of immunohistochemical events ofhealing in elderly normal subjects using 10antibodies relevant to the repair process (filag-grin, keratin, Ulex europeus I, laminen, type IVcollagen, type I collagen, type III collagen, vi-mentin, thrombospondin, and factor VIII) . Thiswound model and timetable will be used forcomparative studies of normal subjects and pa-tients with PVD and DM.

Transcutaneous Oxygen Tension as Predictor of Wound Healing

Frederick A. Matsen III, M.D. ; Ernest M. Burgess, M.D . ; Craig R. Wyss, Ph.D. ; Richard M. Harrington, M.S.Limb Viability Laboratory, University of Washington, Department of Orthopaedics, Seattle, WA 98195Sponsor: VA Rehabilitation Research and Development Service

Purpose—Transcutaneous oxygen tension is amputation surgery.the surface oxygen tension measured with asmall heated sensor sealed to the surface of theskin. Cutaneous heating induces local hyperae-mia in the skin under the sensor and allowsmeasurement of the potential for oxygen deliv-ery in relation to metabolic needs . The oxygenthat diffuses through to the surface of the skinis that delivered in excess of local metabolicneeds. This year our group published reports onthe relation between transcutaneous oxygentension measurements (TcPO2), ankle systolicblood pressure measurements (ABP), and theclinical outcome of both vascular surgery and

Preliminary Results—Both TcPO2 and ABPwere measured before and after 53 vascularprocedures performed to relieve limb-threaten-ing ischemia. In the patients without diabetes,those with post-surgical ABP greater than 75

Hg or TcPO2 greater than 20 mmHgshowed resolution of the clinical symptomswithin 60 days after surgery . All patients fall-ing below these levels underwent subsequentlimb amputation.

Diabetic patients had different results,with many having high ABP in conjunction

235


with low TcPO2 . This was attributed to a highincidence of calcific medial stenosis leading toelevated ABP measurements . The clinical out-come of vascular surgery on diabetic patientswas uncorrelated with the postsurgical ABPand poorly correlated with postsurgical TcPO2 .Those with postsurgical TcPO2 below 20 mmHgall had unfavorable outcomes, but many withhigher values also sustained slow healing ofulcers, persistence of rest pain, and/or limb am-putation.

In the study of 284 amputation surgeries,TcPO2 and ABP were measured within the 2weeks before surgery . The amputations wereperformed at levels selected by clinical criteriaonly. The surgery was classified a success if theamputation wound eventually closed andhealed, even if it required a long period and/orlocal debridement. For both diabetic and non-diabetic patients undergoing either foot orbelow-the-knee amputation, there is a signifi-

cantly increasing risk of amputation failure aspresurgical TcPO2 values decrease . Thirty per-cent of surgeries on limbs with measurementsin the range 11 to 20 mmHg failed; more than60 percent failed on limbs with measurementsless than 10 mmHg. For above-the-knee ampu-tations, presurgical TcPO2 had no predictivevalue in the surgical outcome.

Presurgical ABP appeared to have predic-tive value only for patients without diabetes.Among diabetic patients, calcific medial steno-sis appears to interfere with ABP measure-ments. Even among nondiabetic patients, therewere a large number (greater than 50 percent)with undetectable pulses but sufficient limbblood supply to heal an amputation.

We conclude that presurgical TcPO2 meas-urements are useful as an addition to or re-placement for ABP in evaluating the outcomeof vascular and amputation surgery.

Enhancement of Ulcerated Tissue Healing by Electrical Stimulation

L. Vodovnik, D .Sc . ; A . Stefanovska, Dipl .Eng. ; R . Turk, M .D . ; H. Benko, P .T . ; M. Malezic, Dipl .Eng. ; V . Kosorok, M .D.Electrical Engineering, E . Kardelj University ; University Rehabilitation Institute ; Jozef Stefan Institute, 61000 Ljubljana,YugoslaviaSponsor : National Institute for Handicapped Research, Washington, DC; Slovene Research Community, Ljubljana,Yugoslavia

Progress—Pulsed or DC electrical currentshave been successfully applied in healing indo-lent wounds. In our preliminary experiments,dual-channel stimulation was applied to spinal-cord-injured patients with decubitus ulcers, topostamputation wounds, and to ulcus cruris.After encouraging results on 30 patients, astudy was started on 10 spinal-cord-injured pa-tients with decubitus ulcers, which had devel-oped over a period of several months. Stimula-tion was started 1 week after admission to theRehabilitation Institute . Two channels of bipha-sic stimulation with tetanizing currents wereapplied through four electrodes across thewound. The stimulation parameters were a fre-quency of 40 Hz and a pulse width of 250 micro-seconds . The amplitude (up to 50 mA) was ad-justed to achieve minimal muscle contractionwhen feasible . Stimulation was applied twice

daily for 20 minutes . Once weekly the volumeof the wound was measured by approximatemeasurements of surface and depth, the woundwas photographed, and wound tissue samplesfrom the center, the periphery, and from even-tual pockets were taken for bacterial tests.

The size of all wounds decreased after stim-ulation. The initial healing rate (cm 3/day) de-pends on the initial wound volume (cm 3) . Thehealing rate per 1 cm 3 initial wound volumeranged between 0 .01 and 0 .02 per day . Thus, aninitial wound of 200 cm 3 started to heal with arate of about 2 to 4 cm 3/day, the rate decreas-ing exponentially with time . The time constantof healing is thus between 50 and 100 days.Bacterial tests had previously produced no con-sistent relation between the healing processand changes in the types of bacteria .

236


Acceleration of Fracture Healing Electrica

Carl T . BrightonUniversity of Pennsylvania, Philadelphia, PA 19104Sponsor: National Institutes of Health

Purpose—The object of the proposed researchis to continue investigating the effects of ap-plied electrical fields on the acceleration offracture healing in laboratory animals . Theproposed research was designed : 1) to deter-mine the optimum parameters of applied (exog-enous) electricity for accelerating fracture heal-ing; 2) to determine the role of stress-generated(endogenous) electricity in fracture healing ; and3) to determine the mechanism of electricallyinduced osteogenesis at the cell level.

Methods to be used include the comparisonof the osteogenic response of in vitro fetal rattibia and in vivo healing rabbit fibula to con-stant DC, various pulsed unidirectional electricfields, and various electromagnetic fields . Os-

Fields

teogenesis and bone healing will be evaluatedby incorporation of tritiated thymidine, Ca45,and 35SO4 as well as by maximum resistanceto bending as determined by an Instron TestingMachine. Stress-generated potentials will bemeasured in fracture calluses . Origin of stress-generated potentials will be evaluated by alter-ing collagen in tendon biochemically . Themechanism of action of electrically induced os-teogenesis will be sought by determining: 1)PO2 and pH changes in the vicinity of a cath-ode; 2) changes in surface of cell membrane ; 3)mitochondria release of calcium; 4) cellular pro-liferation and migration; and 5) collagen andproteoglycan biosynthesis and processing.

Biomechanics of Metastatic Defects in Bone

Wilson C. HayesBeth Israel Hospital, Boston, MA 02215Sponsor: National Institutes of Health

Purpose—Advances in the palliative treatmentof patients with established metastatic malig-nancies have not only prolonged patient surviv-al but have also increased the incidence of bonymetastases and subsequent pathological frac-tures. Thus, prevention and effective treatmentof fractures associated with metastatic defectsin bone have become increasingly important as-pects of the care of cancer patients . Unfortu-nately, there are currently available only thecrudest of clinical guidelines for assessing theincreased fracture risk associated with meta-static lesions in bone . Therefore, the appropri-ate time for prophylactic stabilization of im-pending fractures is not known . This investiga-tion will be directed in the long term to the de-velopment of comprehensive biomechanicalguidelines for the orthopaedic assessment andtreatment of metastatic defects in long bones,the proximal femur, and the spine.

A four-phase staged approach will be used .

In Task I, we will conduct retrospective radio-graphic reviews of patients exhibiting metastat-ic lesions in those regions and will determinethe most frequent sites and approximate shapesof the lesions . As part of this task, we will alsodevelop improved diagnostic imaging proce-dures for the description of lesion geometries.In Task II, we will determine in vitro thestrength reductions associated with simulateddefects in long bones, the proximal femur, andthe spine. In Task III, we will use finite ele-ment modeling of defects in these regions toprovide a theoretical framework for interpret-ing the experimental results and assessing thesensitivity of the fracture risk predictors to in-dividual patient variations . In Task IV, we willcombine these findings by developing structuralpredictors of fracture risk for individual pa-tients with particular lesions . Biomechanicalguidelines appropriate for each skeletal regionwill be developed and tested retrospectively in

237


clinical populations . These guidelines should

orthopaedic care of patients with metastatic de-represent a significant improvement in the

fects in bone.

Management of Burn Injuries

J . H. Evans, Ph.D . ; J . M. Courtney, Ph .D. ; J. D . S . Gaylor, Ph .D.Bioengineering Unit, Regional Burns Unit, and Plastic and Oral Surgery Unit, University of Strathclyde, G4 ONSGlasgow, ScotlandSponsor: Dow Corning Corporation

Purpose—Work continues on providing abroad-based better understanding of the burnwound and hypertrophic scar and on improvingthe management of burned patients . Currently,research is directed to assessing burn depth, de-termining mass and energy loss from the burnwound, developing synthetic dressings, in vitroexpansion of autologous epithelium, and thetreatment of hypertrophic scars with siliconegel and pressure garments.

Progress—A thermographic technique, employ-ing a low-cost and convenient pyroelectric Vidi-con camera, is based on previous work using ananimal model along with a mechanicallyscanned detector.

Laboratory and clinical investigations arebeing conducted on novel burn dressings and oncandidate materials including silicone-based

materials and hydro-gels. Consideration isbeing given to mechanical strength and con-formability, mass and energy flux, bacterialcontrol, and drug release. Electromagnetic en-hancement of the rate of expansion of epitheli-um is being studied in vitro to provide a clini-cally acceptable source of autologous skin forthe management of the severely and extensive-ly burned patient.

Preliminary Results—A substantial trial hasdemonstrated that a silicone gel dressing canmake a significant contribution to the rate ofresolution of hypertrophic scars . On many sites,particularly the face and hands, it is more man-ageable and better tolerated than is therapybased on pressure. No significant complicationshave been observed in a trial of 150 patients.

Quantitative Evaluation of Nerve Repair

Vincent R . Lentz, M .D.; Joseph Rosen, M .D .; John Wikswo, Ph.D.; Kenneth Cummins, Ph .D. ; Gordon Abraham, M.S.Rehabilitation Research and Development Center, Veterans Administration Medical Center, Palo Alto, CA 94304Sponsor : VA Rehabilitation Research and Development Service

Purpose—The treatment of peripheral nerveinjuries rarely results in recovery of functionadequately approaching that before the injury.In dealing with these injuries, the physicianfaces two major problems . First, the lack of ob-jective methods for assessing the extent of pe-ripheral nerve injury or the extent of regenera-tion in previously injured nerve compels physi-cians to rely on subjective criteria in makingclinical decisions regarding the management ofthese injuries . Second, present methods of pe-ripheral nerve repair, even those employingmodern microsurgical techniques, rarely result

in regeneration sufficient for full functional re-covery. A number of adjuncts to traditionalrepair methods offer the promise of increasedor directed regeneration, but the effectivenessof these new approaches needs to be assessedobjectively.

The development of a simple method forevaluating the extent of a nerve injury willhelp the physician make objective decisions re-garding the timing and type of care of the in-jured nerve. The use of the appropriate modali-ty for the treatment of nerve injuries will en-hance and speed the regenerative process and

238


will lead to fuller recovery of function.Purely mechanical (suture) methods of

reapproximating peripheral nerves result invery limited functional axonal regeneration. Itis thought that this is due primarily to scar for-mation at the repair site, which prevents proxi-mal axons from crossing the repair site and es-tablishing functional connections. We havebeen studying sutureless repair methods thatuse biodegradable wraps that provide total cir-cumferential alignment of the nerve, block scarinvasion at the repair site, and improve the mi-lieux for axonal regeneration. The preliminaryresults of our early investigation indicate thatnone of the purely mechanical alignment tech-niques hold any significant advantages to re-generation. Regeneration is being assessedusing electrophysiological methods that esti-mate the number and velocity distribution(DCV) of axons both proximal and distal to therepair site. The analysis methods presentlyused require involved mathematical computa-tion and must be performed off line . Such eval-uation methods are currently inappropriate forclinical use by physicians.

This project is based on two fundamentalhypotheses. First, we feel that nerve functionfollowing injury, or axon regeneration followingrepair, can be reliably, quickly, and safely eval-uated by electrophysiological methods that esti-mate the number and health of the axons cross-ing a focal lesion or repair site . For this tool tobe generally applicable, it must be appropriatefor both noninvasive and invasive (intraopera-tive) studies. Second, to enhance nerve regen-eration, in any significant and reliable fashion,modern mechanical repair methods will have tobe coupled with other modalities that intrinsi-cally influence the rate and/or direction ofaxonal regeneration (tropic factors), that im-prove the axonal environment at the repair site(pseudoperineurial barriers), or that decreasethe effect of scar formation.

Progress—We are currently developing asimple technique for characterizing nerve func-tion and regeneration at the repair site . Thismethod estimates the number of axons thatcross a focal lesion (or repair site) and their dis-

tribution of added conduction delays . This ap-proach requires considerably less computationthan DCV analysis. We are also investigatingthe possibility of using magnetic recording and/or stimulating techniques for intraoperativeevaluation of exposed nerve bundles . By em-ploying magnetic coupling for recording, thepractical problems of mechanical nervedamage, stimulus artifact, and operating-roomground loops will be minimized . The resultingsignals, which reflect longitudinal current inthe axons, can be analyzed by any of our meth-ods of nerve evaluation . Initial experiments onmagnetic measurement of action currents fromrat sciatic nerve, performed in cooperation withJohn Wikswo, Ph .D., of Vanderbilt University,have been quite promising . Innovations by Dr.Wikswo permit measurements at room temper-ature using inexpensive equipment.

Preliminary Results—During this project wehave compared the use of a biocompatible, bio-degradable "pseudoperineurial barrier" tostandard microsutures in several primate nerverepair and nerve graft models. In spite of in-traoperative technical problems with the first-generation pseudoperineurial barrier, the ini-tial results from postoperative electrophysiolo-gic assessments of the number of axons havingregenerated across the repair site indicate thatthe barrier does as well as or better than stand-ard suture repair methods.

There is mounting evidence that an auto-immune reaction follows destruction of the per-ineurial blood-nerve barrier during nerveinjury. We have studied the effects of an im-munosuppressive agent, Cyclosporin-A, in de-creasing the initial inflammatory response asso-ciated with nerve injury in a rat model . Prelim-inary histologic results indicate a significantdiminution of inflammatory cells about therepair site . Early electrophysiological studiesindicate some enhancement of regeneration . Incontrast, studies of the effects of the ability ofelectromagnetic fields to enhance regenerationhave not produced significant results.

Encouraging results have been obtainedusing simulated data to test the model for char-acterizing focal conduction delays that occur at

239


the site of nerve injury or repair . The programswill soon be adapted to run on portable evokedpotential equipment, and human intraoperativeassessments of nerve injuries will soon follow.

Several modifications in the neuromagneticsystem have allowed the acquisition of reliable

data from primates and rats. These data indi-cate that the neuromagnetically derived wave-forms can be processed exactly like the elec-tronically recorded counterparts . The develop.ment of an openable clip-on toroid will permithuman studies to begin soon.

Nerve Coupler—Sutureless Peripheral Nerve Repair at the Fascicular Level

Joseph Rosen, M.D.Rehabilitation Research and Development Center, Veterans Administration Medical Center, Palo Alto, CA 94304Sponsor : VA Rehabilitation Research and Development Service

Purpose—Current methods of peripheral nerverepair often result in unsatisfactory restoratonof function of nerves . Factors at the repair sitethat can contribute to these poor results in-clude: 1) poor alignment of the repair site bysuture methods; 2) ingrowth of surroundingscar tissue that blocks the repair site ; and 3)outgrowth of nerve tissue from the repair siteinto surrounding tissue with consequent neur-oma formation.

The importance of our study lies in its po-tential for improving the current state of pe-ripheral nerve repair. Peripheral nerve damageassociated with trauma to bones, tendons, andother structures occurs in approximately 30percent of extremity trauma. Nerve injuriesprevent successful rehabilitation more than anyother form of trauma. Motor dysfunction, lossof tactile discrimination, sympathetic dystro-phy, and postregenerative pain can greatlyimpair self-care and other tasks of daily life,leading to long-term disability . Improved nerverepair would theoretically improve veterans'quality of life following nerve repair.

Present methods of peripheral nerve repairby suture are based on an anatomical approachthat attempts to reestablish continuity of themultiple layers of nervous tissue . In suturerepair, either the epineurial or perineuriallayers of connective tissue are reconnected bysutures to promote healing of the nerve . Insuture repair, the discrete tissue layers are ob-literated by scar tissue at the repair site . An al-ternative to this anatomical approach is a cellu-lar approach involving the construction of anartificial perineurium. The perineurial layer di-

vides the nerve into two distinct cellular envi-ronments . The intrafascicular Schwann cells,axons, and endoneurial fibroblasts compose thenerve's regenerating unit . The cellular compo-nents of the extrafascicular environment areresponsible for the fibrous reaction to injury.

In a number of previous papers, the inves-tigator has developed and described tubes thatreconstruct the perineurium . This suturelessmethod of nerve repair (fascicular tubulization)allows the nerve to reestablish the perineuriumnaturally. Histology and electrophysiology haveshown collagen tubes to be an effective repairmethod in rat monofascicular and cat multifas-cicular models . Bioresorbable polyglycolic acid(PGA) tubes were shown to be successful in therat monofascicular model by qualitative andquantitative histology. No prior methods ofnerve repair have used this fascicular tubuliza-tion technique.

The results of peripheral nerve repair maybe improved by developing a better method toapproximate the transected ends of a nerve.The development of a device that couples theends of the nerve together without suturesmight improve on the poor results of suturerepair. The "nerve coupler" will precisely ap-proximate the ends of a transected nerve fasci-cle. This will improve the alignment at therepair site . The device will also act as a barrierto the ingrowth of scar tissue from outside theperineurium. It will serve to block the out-growth of intrafascicular nerve tissue into theextrafascicular space.

Progress—The first objective of this pilot study

240


was to determine a satisfactory design and ma-terial for the nerve coupler. Prototype plexig-lass couplers and bioresorbable polyglycolic acid(PGA) couplers were studied in the monofasci-cular rat peroneal nerve . These 1- to 3-monthshort-term studies were designed to determinehow effectively the coupler maintained the in-tegrity at the repair site during regeneration.These studies also compared coupler repairwith controlled perineurial suture repairs.Gross and microscopic reactions to coupler ma-terials were evaluated . Repair site organizationwas assessed by histology . The results of thispilot study will be used to modify the design ofthe nerve coupler.

Preliminary Results—Trials of plexiglass cou-plers in two animals demonstrated simple ap-plication of the coupler and excellent nervealignment on gross visual examination . Histo-

logical evaluation showed excellent alignmentand minimal cellular reaction at the couplerrepair site . Studies of 1, 2, and 3 months' dura-tion using PGA couplers again showed simpleapplication of the coupler and excellent grossalignment at the sutureless repair site . Micro-scopic evaluations of 1- to 3-month animals re-vealed fair to excellent organization at the cou-pler repair site and minimal to moderate reac-tion to coupler material.

Future Plans—Our next goal is to obtainVAMC RAG funding to continue our studies ofthe nerve coupler in the rat peroneal nerve . Wewould like to perform long-term studies toevaluate repairs using quantitative histologyand electrophysiology . We would also like toevaluate new designs and new materials for thecoupler in future studies.

Evaluation of Tubular Internal Fixation Plate for Fracture Management

Savio L-Y. Woo; Richard Coutts, M .D . ; David Amiel, Dip .Ing. ; Steven Garfin, M.D . ; Wayne Akeson, M .D.Veterans Administration Medical Center, La Jolla, CA 92103 and University of California, San Diego, CA 92103Sponsor : VA Rehabilitation Research and Development Service

Purpose—The overall objective of this project isto determine whether significant advantagescan be obtained from internal fixation of dia-physeal fractures using plates with improveddesign. Specifically, our goal is to test thenewly developed design criteria . That is, moder-ate bending and torsional plate rigidities areneeded for early immobilization to achieve frac-ture healing without bone shortening and/orangulation, and low plate axial rigidity is desir-able to minimize stress (strain) shielding of theunderlying bone during the postunion remodel-ing process . A new plate with a tubular cross-section is made of stainless steel filled with pol-yethylene. This plate has the aforementioneddesired rigidity characteristics and will be usedin a canine unilateral midshaft femoral osteoto-my mode. A traditional solid stainless steelplate of identical external geometry will beused in a separate series of animals with simi-lar osteotomy. In addition, it is also our goal totest whether such a low axial rigidity plate can

prove to be advantageous to the recovery of fullstructural properties of the underlying bone fol-lowing plate removal.

Progress—At present, the evaluation of thebone lying beneath both the traditional solidstainless steel (control) and the new tubular(experimental) plates is being performed usingX-ray, geometric measurements, histomorpho-metric measurements, and biochemical tech-niques. Specifically, a comparison is being madebetween the actual physical properties of boneand the structural and mechanical propertiesobtained by biomechanical bending and byaxial and torsional testing.

Preliminary Results—Histomorphometric datahave shown that on the average, bone porosityis higher for the control solid stainless steelplate, whereas there is an increase in new bonegrowth and less porosity for bone lying beneaththe experimental hollow plate . This informa-

241


tion correlates well with the bioengineering

more insight into the problem of localized os-tests performed . With further comparison and teopenia that occurs with the use of internalstudy of these test results, we hope to provide

fixation plate systems.

Biomechanical Considerations of Metal and Composite Materialsfor Bone Fracture Fixation Plates

Savio L-Y. Woo ; Wayne H. Akeson; Richard D . CouttsUniversity of California, San Diego and San Diego Veterans Administration Medical Center, La Jolla, CA 92103Sponsor : VA Rehabilitation Research and Development Service

Progress—This study, together with others, re-futes the necessity of extremely high platebending and torsional rigidities (absolute immo-bilization of the fractured bone ends) for union.Plates with significantly less rigidities, and ap-plied without "compression", were able toachieve equivalent and better bone healingwith the aid of callus formation. The results ofthis series of experiments confirm that osteonalbone union is slow and mechanically inferior tocallus union.

The design criteria developed by our labo-ratory—that is, moderate bending and torsionalrigidities (to provide adequate but not "rigid"

fixation of fractured bone for formation of per-iosteal callus and for union) together with thelow plate axial rigidity (to allow the healedbone to share a larger portion of the physiologi-cal stresses to reduce the instances of stress-protection osteopenia)—may be an ideal com-promise between investigators who advocaterigid internal fixation systems and those whoadvocate flexible ones . A tubular cross-sectionalstainless steel plate may be an optimal designfor such a less rigid fixation plate system.Other materials and designs following theseguidelines also should be pursued .

Journal of Rehabilitation Research and Development

Documents

Transcript of Journal of Rehabilitation Research and Development