Aetiology of pressure ulcers

Sandra LoerakkerOctober, 2007BMTE 07.39

Promotor: prof. dr. ir. F.P.T. Baaijens

Coach: dr. ir. C.W.J. Oomens

Eindhoven University of TechnologyDepartment of Biomedical EngineeringSection Materials TechnologyDivision Biomechanics and Tissue Engineering

Contents

1 General information 21.1 Prevalence and costs of pressure ulcers . . . . . . . . . . . . . . . . . . . 21.2 Classification systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21.3 Risk assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

2 Mechanical loading 62.1 Pressure, shear and friction . . . . . . . . . . . . . . . . . . . . . . . . . 62.2 Internal stress state . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72.3 Magnitude and duration of loading . . . . . . . . . . . . . . . . . . . . . 9

3 Damage pathways 113.1 Localized ischemia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113.2 Ischemia/reperfusion injury . . . . . . . . . . . . . . . . . . . . . . . . . 133.3 Impaired lymphatic drainage . . . . . . . . . . . . . . . . . . . . . . . . 163.4 Sustained deformation of cells . . . . . . . . . . . . . . . . . . . . . . . . 18

4 Discussion 214.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214.2 Focus questions for our research . . . . . . . . . . . . . . . . . . . . . . . 22

Bibliography 24

1

Chapter 1

General information

1.1 Prevalence and costs of pressure ulcers

Pressure ulcers are localized areas of tissue necrosis that tend to develop when soft tissueis compressed between a bony prominence and an external surface for a prolonged periodof time [58]. They occur in situations where people are subjected to sustained mechanicalloads and are particularly common in subjects who are bedridden or wheelchair-bound.Prevalence figures remain very high; e.g. in a recent Dutch prevalence study it wasshown that the prevalence of pressure ulcers was 15.4% in university hospitals, 11.8% ingeneral hospitals, 18.3% in nursing homes and 8.4% in home care [47]. Especially spinalcord injury (SCI) patients are at risk for developing pressure ulcers, since prevalencerates for this group are much higher. For SCI patients, the occurrence of pressure ulcersis among the most common long-term secondary medical complications [50]. Accordingto Dinsdale, the incidence of pressure ulcers in paraplegic and quadriplegic patientsranges from 25% to 85% [24]. In addition, Garber and Rintala found that a large part(56%) of the ulcers in SCI patients are grade IV pressure ulcers (most severe, see section1.2) [27]. These ulcers usually start in deep tissues over bony prominences and are hardto detect before they reach the skin surface.

Besides the fact that pressure ulcers are very painful for patients themselves, theyalso represent a huge financial burden for health care in Western countries. In theNetherlands, more than 1% of the total health care budget is spent on prevention andtreatment of pressure ulcers or the prolonged hospital stay once a pressure ulcer hasdeveloped. Here, pressure ulcers are the third costliest disorder, after cancer and car-diovascular diseases [55].

1.2 Classification systems

There are different systems to classify pressure ulcers. The system adopted by the Eu-ropean Pressure Ulcer Advisory Panel (EPUAP) consists of four grades in which eachgrade is defined by the anatomic limit of tissue damage, see table 1.1 [23] and figure1.1. This classification system can be used to evaluate the anatomic depth of tissue

2

Chapter 1. General information 1.2. Classification systems

Table 1.1: Classification system of the EPUAP [23].

Grade Short description DefinitionGrade I Non-blanchable erythema of

intact skinA discolouration of the skin,warmth, oedema, induration orhardness may also be used as indi-cators, particularly on individualswith darker skin.

Grade II Abrasion or blister A partial loss in the thickness of theskin involving epidermis, dermis, orboth.

Grade III Superficial ulcer A full loss in the thickness of theskin involving damage necrosis ofsubcutaneous tissue that may ex-tend down to, but not through, un-derlying fascia.

Grade IV Deep ulcer An extensive destruction, tissuenecrosis, or damage to muscle,bone, or supporting structures withor without full loss in the thicknessof the skin.

destruction, but is also erroneously used in reverse order to monitor the healing processof an ulcer [49, 59]. This is not correct, for as stage IV ulcers granulate progressivelyto a shallower depth, they do not replace the structural layers of muscle, subcutaneousfat, and dermis before they reepithelialize [49]. Furthermore, the inter- and intraraterreliability of the EPUAP classification system appeared both to be low for nurses. Mis-classification of pressure ulcers can then lead to inadequate preventive and therapeuticmeasures and suboptimal nursing care [23]. Besides the fact that the system cannotbe used for monitoring wound healing, it also focusses only on visible signs of skin orunderlying tissues, so ulcers that initiate in deeper tissues, e.g. near bony prominences,are not recognized until they reach the skin and instantly become a grade IV ulcer [25].Ankrom et al. also mentioned that the published literature did not offer consensus onhow to describe, diagnose, or treat pressure-related deep tissue injury under intact skin[2]. Recently, the National Pressure Ulcer Advisory Panel (NPUAP) updated their clas-sification system and also added deep tissue injury as a distinct pressure ulcer in thisupdated system [6].

3

Chapter 1. General information 1.3. Risk assessment

Figure 1.1: Schematic representation of the four grades of pressure ulcers [1]. A grade I ulcer ischaracterized by a discoloration of the skin, a grade II ulcer by partial loss in thickness of the skin, agrade III ulcer by full loss in thickness of the skin, and a grade IV ulcer by extensive destruction ofmuscle, bone, or supporting structures with or without full loss in the thickness of the skin.

1.3 Risk assessment

There are two major factors associated with the risk of developing pressure ulcers: theamount and duration of exposure to pressure and the ability of the tissue to toleratethe pressure [5], see figure 1.2. The primary cause of pressure ulcer development isthe exposure to pressure; without a mechanical load no pressure ulcer will develop.However, the ability of the tissue to withstand mechanical loading determines whetheror not a person will develop a pressure ulcer due to a certain loading. The exposure topressure is influenced by the mobility, activity and sensory perception of a patient andtissue tolerance for pressure is influenced by intrinsic and extrinsic factors [5]. Intrinsicfactors are related to the individual, e.g. age, smoking, incontinence, weight and bodytemperature. Extrinsic factors are related to the environment, e.g. contact surface andtemperature and humidity of the environment.

In clinical practice, these risk factors are used in risk assessment scales to identifypatients at risk of developing pressure ulcers. Many different risk assessment scales exist,but the value of these scales in predicting the risk of pressure ulcer development remainsquestionable. For example, Schoonhoven et al. tested three commonly used scales (theNorton, Braden and Waterlow scales) and concluded that the use of the outcomes ofthese risk assessment scales to decide on preventive measures leads to ineffective andinefficient treatment for most patients [75]. Moreover, these scales cannot be used forSCI patients, since they are always classified as high risk patients. Still, they may stayfree of ulcers for a long time, and suddenly develop an ulcer in a very short period.

4

Chapter 1. General information 1.3. Risk assessment

Figure 1.2: Conceptual schema which accounts for the relative contributions of the duration andintensity of pressure and the tolerance for pressure.

5

Chapter 2

Mechanical loading

2.1 Pressure, shear and friction

Mechanical loading that can lead to the development of pressure ulcers not only consistsof pressure; shear and friction are also mentioned to be important [24, 69]. It is believedthat shear and friction mainly cause superficial ulcers, while pressure usually leads todeep tissue destruction, especially in soft tissues over bony prominences [13]. In practicehowever, it is difficult to have conditions of pure pressure, and impossible to developpure shear [3]. For example, the application of pressure by means of a piston leads to ashear stress near the piston edge [4], and the absence of pressure during the applicationof shear will promote slip [3].

Not all kinds of mechanical load are equally damaging for biological tissues. Hydro-static pressure does not have a large influence on the development of pressure ulcers,since divers can work at large depths without developing pressure ulcers [8, 53, 81].Localized pressure on the other hand, causes deformation of the tissues and blockage ofblood vessels, and is therefore far more damaging [20]. Furthermore, Nola and Vistnesfound that, by applying pressure on skin and muscle tissue over bone in rats, that moreulceration appeared in the muscle layer than in the skin [54]. They therefore concludedthat muscle was more sensitive to pressure than skin and subcutaneous tissue. Salcidoet al. also observed that lesions occurred first in the muscle rather than in the skin whenapplying pressure to the trochanteric region of rats [74]. However, instead of being moresensitive to mechanical loading than skin, the muscle tissue in these studies could alsobe subjected to higher loadings than the superficial layers, see also section 2.2. Thiswould also explain the earlier occurrence of ulceration in muscle tissue.

An applied shearing load does not lead to large tissue destruction, e.g. Bennett andLee found that externally applied pressure was approximately twice as effective as ap-plied shear in reducing pulsatile blood flow [3]. However, the combination of pressure andshear is very damaging, e.g. at a sufficiently high level of shear, the pressure necessaryto produce occlusion was only half that when little shear was present [3]. Furthermore,Dinsdale found that friction also increases the susceptibility to skin ulceration, sincethe presence of friction decreased the amount of pressure necessary to produce ulcers in

6

Chapter 2. Mechanical loading 2.2. Internal stress state

paraplegic swine [24].The amount and distribution of external pressure, shear and friction during sitting or

lying is influenced by several factors. Firstly, the body configuration has a large influenceon the applied mechanical loading [35]. In addition, the properties of the cushion ormattress material play an important role. Finally, the individual characteristics alsoaffect the kind and distribution of mechanical loading. For example, in sitting on a hard,slick seat, paraplegic, geriatric, and ill subjects develop roughly three times the medianshear load experienced by normal subjects in skin lateral to the ischial tuberosities [3].Furthermore, by investigating the effects of seated posture and body orientation on thepressure distribution and shear forces acting at the body-seat interface, Hobson foundthat individuals with SCI have significantly higher maximum pressures than nondisabledsubjects in all postures that were studied [35].

2.2 Internal stress state

In clinical practice, the interface pressures between the body and the supporting surfaceare measured to assess the mechanical load that is applied to the body and thereby therisk of developing pressure ulcers. Interface pressures are however not a good measure toestimate the risk for deep pressure ulcers, since they do not provide information aboutthe stress state in deeper tissue layers. For example, it was shown by numerical models[20, 31, 45, 46, 57, 80, 82, 86] and physical models [17, 67] that stresses and strains arenot uniform within the tissues. Moreover, the nature of the external loading does notresult in the same kind of loading within the tissues, which is illustrated in figure 2.1.

Figure 2.1: Schematic drawing of a uniaxial compression test. Solid and dotted lines show the unde-formed and deformed configurations, respectively [56].

In this figure, a uniform axial compression is applied to an object. When smallrectangular part the object is observed as in situation a, then the deformation of thesmall part is similar to the macroscopic deformation, and the object only experiencescompression. However, if the small part is rotated as in situation b, then it experiencescompression as well as shear. In fact, for each material point a configuration can be

7

Chapter 2. Mechanical loading 2.2. Internal stress state

found in which only compressive or tensile stresses and strains are present, as well asa configuration in which the maximum shear strain is reached. Thus, although themacroscopic deformation was compressive, it results in pressure and shear in the tissue[56]. Furthermore, since hydrostatic pressure is relatively harmless to biological tissues,the harmfull stress is presumably a form of shear [20].

Several numerical studies in which soft tissues are compressed, showed that stressesand strains in the muscle layer adjacent to bony prominences are much higher thanthose near the surface [31, 45, 46, 57], see also figure 2.2. In addition, the mechanicalstate of the tissue can change over time [48]. Moreover, Linder-Ganz et al. showed thatstresses and strains in muscle are less homogeneous than in fat, and that muscle tissuewas also subjected to tensile stress where fat was not [46]. Furthermore, the study ofOomens et al. showed that the highest shear strains occurred in fat [57]. However, Sunet al. reported that the maximum shear stresses occurred in muscle tissue and withinskin, and that they were significantly higher than in fat [80].

(a) (b)

Figure 2.2: Two examples of finite element models that show that stresses and strains are higher nearbony prominences than near the surface. a) Deformed mesh of buttock model of Oomens et al. withvon Mises stress [57]. b) Principle compressive strains in deep tissues under the ischial tuberosities of ahealthy subject, modeled by Linder-Ganz et al. [46].

The internal stress state in the tissues depends on several factors. Firstly, the cushionproperties have an influence on the stresses and strains within the tissues [57, 67]. Forexample, Oomens et al. mentioned that a change in cushion properties mainly influencesthe shear strains in the fat layer [57]. Furthermore, Linder-Ganz and Gefen showed thata change in mattress stiffness led to larger changes in contact pressure and shear thanin peak internal stresses [45]. Secondly, the mechanical properties of the tissues alsoaffect the stress and strain distribution. For example, a higher Young’s modulus inthe model of Chow and Odell resulted in smaller deformations and a more uniformstress distribution [20]. Mechanical properties of tissues can change when cell becomesdamaged. It was found by Gefen et al. and Linder-Ganz and Gefen that damaged

8

Chapter 2. Mechanical loading 2.3. Magnitude and duration of loading

muscle tissue becomes stiffer [31, 45]. This increase in muscle stiffness leads to higherstresses in the injured muscle parts in the numerical model of Gefen et al., and thesestresses were then transferred to adjacent tissue that was not yet injured [31]. Finally,the shape of a bony prominence also influences the internal stress state. This was shownby the physical buttock model of Candadai and Reddy, which consists of a PVC gel (softtissue) molded around a wooden core (bone), in which two shapes of bone were tested[17]. Furthermore, a theoretical analysis of Gefen showed that the size of the injuredmuscle area and the time for injury onset depended on the radius of curvature of thebony prominence [30].

It is clear that the internal stress state is nonuniform in the tissues that are subjectedto a mechanical load. However, the internal distribution of stresses and strains does notdirectly determine whether or not a pressure ulcer will develop at certain locations. Thisalso depends upon the load threshold of the individual tissues.

2.3 Magnitude and duration of loading

Besides the mechanical loading of tissues, the time of exposure also plays an importantrole in the development of pressure ulcers. The influence of both factors was studied inanimal models. In 1942, Groth applied pressure to the posterior ischii of rabbits, andfound that an inverse relationship existed between the minimum pressure and minimumduration required to produce a pressure ulcer [33]. Husain confirmed the shape of thisrelationship by applying pressures of 100 to 600 mmHg to the legs of normal rats for1 to 6 hours [37]. In 1959, Kosiak applied pressures ranging from 100 to 550 mmHgfor periods of 1 to 12 hours to the femoral trochanter and ischial tuberosity of dogs.This also revealed an inverse relationship between the minimum pressure and time thatlead to ulceration of tissue [39]. In 1981, Daniel et al. subjected the greater femoraltrochanter of pigs to pressures of 30 to 1000 mmHg for periods of 2 to 18 hours, whichalso resulted in a hyperbolic pressure-time curve [21]. In 1976, a similar hyperbolic curvewas produced for humans by Reswick and Rogers, who used comments of physicians,nurses and therapists, pressure measurements, and controlled tests on volunteers [70].A large variation however exists between the threshold values found in different studies,see figure 2.3(a). This is due to diversity in experimental conditions, animal models,loading methods and locations of load application [77].

The shape of the curve was also determined by Sacks, who used dimensional analysisto determine a relationship between the required pressure to produce an ulcer, and thephysical properties of the tissue, the blood flow through it, and the time of exposure [73].This analysis showed that the pressure was a function of t−4/3. If a linear relation wasassumed between the pressure and t−4/3, a fit of the equation to the data of Kosiak [39]and Reswick and Rogers [70] resulted in a good agreement for both experiments [73].

9

Chapter 2. Mechanical loading 2.3. Magnitude and duration of loading

(a) (b)

Figure 2.3: a) Pressure versus time curves determined by different studies. Time/pressure combinationsabove the curves result in tissue breakdown [77]. b) Adapted pressure versus time curve determined byLinder-Ganz et al. [44].

The hyperbolic pressure-time curve implies that loads that are applied for very shorttimes would not lead to tissue damage. To investigate if this was true, Linder-Ganz etal. also applied varying loads for short durations and showed that high loads could leadto damage instantly [44]. The pressure-time curve as proposed previously is thereforeprobably not complete, and shows a more sigmoidal shape when short duration dataare also included, see figure 2.3(b). This means that for short exposures (less than 1hour) and also for long exposures (over 2 hours) the magnitude of the pressure is themost important factor for causing cell death; the exposure time has little or no effect.However, for the intermediate exposures (between 1 and 2 hours) pressure and timeboth are important factors [44].

The pressure-time relationship was also investigated for paraplegic animals. Groth[33] and Kosiak [40] studied the response of paraplegic animals when compared to nor-mals. Both found no significant differences in susceptibility. Daniel et al. howevernoted that the animal model of Groth was probably experimented too soon followingtransection of the spinal cord, thereby precluding the tissue atrophy characteristic of theclinical condition [22]. Probably this also holds for the animal model of Kosiak. Danielet al. therefore waited six weeks after transection of the spinal cords before using hisparaplegic pigs. They found that the pressure-time curve for paraplegic pigs was alsohyperbolic, but the magnitude of the variables was greatly reduced. Spinal-transectedanimals would therefore develop pressure sores at lower magnitudes of pressure and aftera shorter time than normal animals. The difference between the results was attributedto tissue atrophy, which would lead to increased interface pressures between the supportsurface and the soft tissues overlying bony prominences [22].

10

Chapter 3

Damage pathways

Although it is known that sustained mechanical loading is the primary cause of pressureulcers, the underlying pathways whereby mechanical loading leads to tissue breakdownare hardly understood. At the moment theories involve:

• Localized ischemia;

• Ischemia/reperfusion injury;

• Impaired lymphatic drainage;

• Sustained deformation of cells.

These four theories will be described in the following sections.

3.1 Localized ischemia

Ischemia is often considered to be the most important factor in the development ofpressure ulcers [21, 39, 40]. When a mechanical load is applied to the body, bloodvessels in the loaded tissue can collapse, thereby stopping blood flow and depriving thetissue of its oxygen and nutrients supply. In 1930, Landis studied the blood pressurewithin a capillary loop in healthy human fingernail beds using a microinjection technique,and found that an average pressure of 32 mmHg was sufficient to close the capillaries[43]. The validity of this value should however be questioned, since the capillaries werecannulated , which could have resulted in lower pressure readings as the bleeding vesselswere not full or enclosed. Furthermore, capillary pressures can rise during mechanicalloading due to autoregulation, in which case higher external pressures are needed to closethe blood vessels [3, 81]. Finally, capillary closure depends on local pressure gradientsacross the vessel wall and not just on interface pressures [13]. Still, the capillary closingpressure found in Landis’ study is frequently used in clinical practice as a threshold fortissue damage, where care is taken to avoid interface pressures higher than 32 mmHg.

External pressures that are high enough to close blood vessels will lead to ischemia,which initiates a series of chemical and pH imbalances, accompanied by enhanced gener-ation of injurious free radical species. The damage produced by short periods of ischemia

11

Chapter 3. Damage pathways 3.1. Localized ischemia

tends to be reversible if the circulation is restored, but cells subjected to long episodesof ischemia become irreversibly damaged and die [71]. In general, muscle appears to betolerant of ischemia for up to 4 hours, fat up to 13 hours, and skin up to 24 hours atnormothermia [7]. Ceelen et al. developed a finite element model that describes defor-mation, diffusion of oxygen and damage in a cross-section of skeletal muscle tissue [18].They investigated whether the cessation of oxygen consumption in dead cells could bebeneficial for the remaining cells in the tissue, and if a change in mechanical propertiesupon cell death had an influence on damage development in other cells. The resultsshowed that when cells stopped consuming oxygen after cell death, further cell death isdelayed or even prevented. This effect is larger after more cells have died, because theamount of extra oxygen available for the other cells increases, and is more pronouncedat lower compression levels, since at high compression levels there is already hardly anyoxygen available, and therefore the little amount of extra oxygen hardly has any effect.A change in mechanical properties of cells upon cell death, in this study it was assumedthat dead cells were less stiff, eventually led to delayed cell death at high compressionlevels. At low compression levels or when only a few cells had already died, no changeof mechanical properties of dead cells resulted in slower cell death.

During ischemia, cells have to switch from aerobic to anaerobic metabolism due tothe lack of oxygen. Anaerobic metabolism results in the generation of lactic acid whichleads to a decrease in intracellular pH. The depletion of glucose and accumulation oflactic acid which decreases the intracellular pH both lead to cell death. For example,Gawlitta et al. showed that glucose deprivation and acidification in C2C12 murinemyoblasts resulted in increased cell death [29].

Figure 3.1: Schematic overview of the changes that occur during ischemia [71].

12

Chapter 3. Damage pathways 3.2. Ischemia/reperfusion injury

Intracellular ion concentrations also change during ischemia. Due to the lack of ATP,the Na+/K+ ion exchanges becomes inactive, which subsequently leads to the activationof the Na+/H+ ion exchanger. When intracellular acidosis threatens, it pumps H+ outof the cell in exchange for Na+ to maintain proper intracellular pH. The increase inintracellular Na+ then leads to activation of the Na+/Ca2+ ion exchanger, enhancingcalcium entry. Since calcium pumps are ATP dependent and cannot pump the excessCa2+ out of the cell anymore, calcium accumulates in the cells. High calcium concentra-tions in ischemic cells activate phospholipase A2, leading to degradation of membranephospholipids and the consequent release of free fatty acids and lysophospholipids. Theseare both potent mediators of inflammation and lysophospholipids also act as detergentsthat solubilize cell membranes. The combination of electrolyte imbalance and increasedpermeability of the cell membrane causes cell swelling [71]. Figure 3.1 gives a schematicoverview of the changes that occur during ischemia.

The rise in intracellular calcium concentration during ischemia was confirmed byBoffi et al. who examined intracellular calcium levels in mouse-derived C2C12 myotubeswith inhibition of glycolytic and oxidative metabolism as ischemic condition [9]. On theother hand, Smith et al. found that the total tissue calcium contents of preischemicand end ischemic canine gracilis muscle were similar and started to increase after theischemic period during reperfusion. They therefore concluded that calcium influx is afeature of reperfusion rather than the ischemic interval [76]. Furthermore, when the cellmembrane ionic pumps become depressed, more ions enter the cells than leave them.This leads to osmosis of water into the cells and thus causes intracellular edema [34, 51].

Although ischemia will eventually lead to tissue death, it is probably not the onlycause of pressure ulcers. One reason for this is that pressure ulcers can develop withintwo hours, whereas tissues can withstand ischemia for longer times. Furthermore, is-chemia alone fails to explain why higher tissue pressures can create ulcers after a shortperiod of ischemia, whereas lower pressures, which still create an ischemic state, needlonger periods to cause the identical lesion [60].

3.2 Ischemia/reperfusion injury

Tissue can also become damaged when the blood flow in the tissue is restored afteran ischemic period. Ischemia and reperfusion of skeletal muscle can trigger a seriesof deleterious phenomena in tissue, such as cell edema, increased permeability in themicrocirculation, induction of the no-reflow phenomenon, free oxygen radical production,electrolytic changes in mitochondria, cytosolic calcium overload, and degradation ofmembrane phospholipids [32].

After a relatively short ischemic period, the blood flow is temporarily higher thanunder normal circumstances. This increased blood flow after occlusion is called reactivehyperemia. It is a consequence of a local regulatory mechanism whereby the arteriolesare dilated and the resistance to blood flow is reduced [51]. Ikebe et al. investigatedthe relationship between the duration of ischemia and the subsequent reperfusion bloodflow in rats [38]. They found increased reperfusion blood flow after 90 minutes and

13

Chapter 3. Damage pathways 3.2. Ischemia/reperfusion injury

3 hours of ischemia, but after 6 hours of ischemia there was no significant increase inpostischemic blood flow. This is called the “no-reflow phenomenon” and is probablycaused by cellular swelling, thrombosis, and white cell plugging in capillaries, whichincreases the resistance in the microcirculation [85]. The incidence of no flow dependson the severity of both the ischemic insult and the subsequent reperfusion injury [38].

During reperfusion oxygen is provided to the tissues, which combines with the freeradical species generated during ischemia to form reactive oxygen species (ROS) [71].When present in excess they damage the endothelium, attracting platelets and granulo-cytes, stimulating stasis of blood flow and thrombosis, further decreasing blood flow andthereby stimulating the development of tissue necrosis [36]. The superoxide anion (O−

2 )is produced principally by leaks in mitochondrial electron transport or as part of theinflammatory response, see figure 3.2. In the case of phagocytic inflammatory cells, acti-vation of a plasma membrane oxidase produces O−

2 , which is then converted to hydrogenperoxide H2O2 and eventually to other ROS. H2O2 is also produced directly by a num-ber of oxidases in cytoplasmic peroxisomes. By itself, H2O2 is not particularly injurious,but when produced in excess, it is converted to hydroxyl radicals (•OH). In neutrophils,myeloperoxidase transforms H2O2 to the potent radical hypochlorite (OCl−), which islethal for microorganisms and cells.

Figure 3.2: Mechanisms by which reactive oxygen species are generated from molecular oxygen andthen detoxified by cellular enzymes. CoQ = coenzyme Q; GPX = glutathione peroxidase; SOD =superoxide dismutase. [71].

Hydroxyl radicals are formed by (1) the radiolysis of water, (2) the reaction of H2O2

with ferrous iron (Fenton reaction), and (3) the reaction of O−2 with H2O2 (Haber-Weiss

reaction). The hydroxyl radical is the most reactive molecule, and there are severalmechanisms by which it can damage macromolecules [71]:

• Lipid peroxidation: The hydroxyl radical removes a hydrogen atom from theunsaturated fatty acids of membrane phospholipids, a process that forms a free

14

Chapter 3. Damage pathways 3.2. Ischemia/reperfusion injury

lipid radical, which subsequently initiates a chain reaction. The destruction of theunsaturated fatty acids of phospholipids results in a loss of membrane integrity.

• Protein interactions: Hydroxyl radicals may also attack proteins. As a resultof oxidative damage, proteins eventually undergo degradation.

• DNA damage: DNA is an important target of hydroxyl radicals. If the oxidativedamage is sufficiently extensive, the cell dies.

The major enzymes that convert ROS to less reactive molecules are superoxide dis-mutase (SOD), catalase and glutathione peroxidase. SOD converts O−

2 to H2O2 and O2.Catalase in one of the two enzymes that complete the dissolution of O2 by eliminatingH2O2 and therefore its potential conversion to •OH. Glutathione peroxidase (GPX) cat-alyzes the reduction of H2O2 and lipid peroxides in mitochondria and the cytosol, seefigure 3.2. GPX uses reduced glutathione (GSH) as a cofactor, producing two moleculesof oxidized glutathione (GSSG) for every molecule of H2O2 reduced to water. GSSGis reduced to GSH by glutathione reductase [71]. In tissues undergoing oxidative stressthe amounts of these enzymes decrease, so during reperfusion the oxygen free radicalsare not buffered, or are buffered to a lesser degree [36]. Furthermore, retinoids, theprecursors of vitamin A, are lipid soluble and function as chain-breaking antioxidants.Vitamin C is water soluble and reacts directly with O2, •OH, and some products of lipidperoxidation. Vitamin E blocks free-radical chain reactions. Since it is fat soluble, itexerts its activity in lipid membranes, protecting them against lipid peroxidation [71].Houwing et al. showed that pretreatment with vitamin E prevents damage caused bypressure applied to the femoral trochanters of pigs to a large extent. Vitamin E didnot prevent oxidative stress during pressure application, but it did prevent the excessproduction of oxygen free radicals and hydrogen peroxide during reperfusion [36].

It is hypothesized that the reperfusion phase after an ischemic period is more dam-aging to tissues than ischemia itself. Peirce et al. subjected rats to different numbersof ischemia/reperfusion (I/R) cycles (2 hours of ischemia and 0.5 hour of reperfusion),varied the duration of the ischemic insult, and compared ischemia-induced injury withI/R-injury [65]. They found that tissue injury increased with an increasing number ofI/R cycles and duration of ischemia, and moreover that when the total extent of is-chemia was constant, a greater number of reperfusion events during that period resultedin increased tissue damage. It was therefore concluded that the reperfusion phase ofthe I/R cycle is an important component of the total injury produced. Tsuji et al. sub-jected mice to two different loading protocols and used the functional capillary densityas a measure for tissue damage [83]. The total duration of compression was equal forboth groups, but the first group received four cycles of 2 hours compression and 1 hourrelease, while the second group was exposed to a continuous compression of 8 hours.The results showed that the cyclic compression-release procedure significantly decreasedfunctional capillary density as compared to continuous compression, from which theyconcluded that repetition of the ischemia-reperfusion cycle was more damaging than asingle prolonged ischemic insult, see figure 3.3(a). Furthermore, when applying pressureto the skin immediately above the greater femoral trochanters of pigs, Houwing et al.

15

Chapter 3. Damage pathways 3.3. Impaired lymphatic drainage

found that early signs of damage in the muscles and subcutaneous tissue appeared onlyafter a reperfusion time of one to two hours and not during the ischemic period [36].

Since I/R injury is more damaging than ischemia alone, it was investigated if grad-ual reperfusion was better than complete reperfusion at once. Durrani et al. isolatedthe left renal artery and vein of rats [26]. Microclamps were applied for 45 minutesand released at once in one groups and gradually in another group. Although therewere no significant differences between both groups in the malondialdehyde (MDA) andmyeloperoxidase (MPA) levels, which are indicators of lipid peroxidation and polymor-phonuclear infiltration, respectively, histopathologic scoring showed less tissue damagein the gradual reperfusion group when compared to the conventional reperfusion group.Ikebe et al. regulated the postischemic blood flow in rats by the administration of L-NMMA (nitric oxide synthase inhibitor) to investigate alterations in the viability andcontractile function of the skeletal muscle [38]. They found that muscle viability of theL-NMMA-treated group was significantly better, which suggests that excessive bloodflow during reperfusion deteriorates skeletal muscle contractile function. Unal et al. in-vestigated the gradual reperfusion in a rat hind limb model and tried to elucidate itspotential beneficial effect [84]. Their results showed that the MDA and MPO levels weresignificantly larger in the conventional reperfusion group than in the control and gradualreperfusion groups. There was no significant difference between the levels in the controland gradual reperfusion groups. Inflammatory cell infiltration and loss of striation ofthe muscle were also noticeably less in the gradual reperfusion group than those of theconventional reperfusion group.

(a) (b)

Figure 3.3: Effects of reperfusion. a) Percentage of microcirculatory injury is higher after four ischemia-reperfusion cycles than after continuous compression with the same duration of ischemia [83]. b) Thetissue damage of group II rats (conventional reperfusion) is significantly higher than the damage ingroup III rats (gradual reperfusion). Group I was a control group (without ischemia) [84].

3.3 Impaired lymphatic drainage

Ischemia and I/R injury cannot be the only factors contributing to the developmentof pressure ulcers. If oxygen would be the only factor, then all pressure intensitiesthat cause capillary closure should produce ulceration in the same duration, which is

16

Chapter 3. Damage pathways 3.3. Impaired lymphatic drainage

in contrast to the measured inverse relation between pressure and duration required toproduce ulcers [66]. It is believed that the lymphatic system can also play an importantrole [42, 41, 52, 66]. The lymphatic system functions as an “overflow mechanism” toreturn excess proteins and excess fluid volume from the tissue spaces to the circulation.Therefore, it plays a central role in controlling: (1) the concentration of proteins inthe interstitial fluids, (2) the volume of interstitial fluid, and (3) the interstitial fluidpressure [34].

During mechanical loading, the lymphatics can collapse, thereby obstructing lymphflow. This leads to an accumulation of waste products. Miller and Seale investigated therelation between the external pressure and lymph clearance in dogs [52]. They found anonlinear rise in lymph flow with increasing pressure until a critical closing pressure wasreached. At this point, the flow ceased. The increase in flow with increasing externalpressure was believed to be caused by the increased translymphatic pressure drop atlarger external pressures, which promotes increasing flow into the lymphatics. Therefore,as pressure increases, so does flow, until the external pressure collapses the vessel. Atan applied pressure of 60 mmHg, some animals showed relatively large clearance levelsindicating enhanced lymph flow. Others showed reduced lymph flow indicating the onsetof vessel closure, which implied that the critical closing pressure might be at or near 60mmHg, see figure 3.4.

Figure 3.4: Lymph flow per unit tissue volume as a function of applied pressure. Lymph flow increaseswith pressure until a critical closing pressure is reached. Some data at 60 mmHg show vessel closure andsome show enhanced flow, indicating that the critical closing pressure may be at or near 60 mmHg [52].

It is also hypothesized that the application of mechanical loads impairs the con-tractility of the lymphatic vessels. Krouskop et al. postulated that hypoxia caused byexternal loading leads to damage of lymphatic smooth muscle which in turn results inthe loss of lymph motility and impaired lymph flow [42, 41].

Another hypothesis implies that when interstitial fluid is squeezed out of a region,direct contact of cells (fibroblasts) may induce contact stresses. These may in turn causecontact inhibition or even rupture and interrupt collagen synthesis. Furthermore, if theexternal pressure is removed, the interstitial fluid pressure might become small enough

17

Chapter 3. Damage pathways 3.4. Sustained deformation of cells

to cause capillary bursting, edema and lymphatic damage [66].The lymphatic flow was also studied by means of mathematical models. For example,

Reddy et al. offered a phenomenological theory to explain the formation of pressureulcers [66]. They considered a cylinder of tissue and assumed that the flow in connectivetissue can be described with Darcy’s law. They found that the product of the interstitialfluid pressure and time was constant for a reduction in interstitial fluid volume, whichmight lead to tissue damage. Since this inverse relationship is in correspondence withother studies in which the relationship between pressure intensity and load durationrequired to create ulcers was investigated, it was concluded that interstitial fluid flowcould play a significant role in the mechanisms responsible for this inverse relationship,and therefore also in tissue breakdown.

In a later study, Reddy and Patel developed a mathematical model to simulatelymph flow through the terminal lymphatics under different physiological conditions[68]. Blood pressure and protein concentration in the capillary formed the upstreamboundary conditions and the pressure in the adjacent contractile lymphatic was takenas the downstream boundary condition. Results showed that the flow increased withincreasing stiffness of the terminal lymphatic vessel wall, increasing stiffness of the an-choring filaments, and increasing hydraulic conductivity of the terminal lymphatic vesselwall and blood capillary wall. Also, the flow through the terminal lymphatics increasedduring edema, which was simulated by increasing the capillary pressure. Furthermore,the results showed that the flow through the terminal lymphatics is due to periodic fluc-tuations in interstitial fluid pressure and due to the suction mechanisms of the adjacentcontractile lymphatics.

3.4 Sustained deformation of cells

Recently, it was hypothesized that sustained deformation of cells could also be a causeof pressure ulcers [12, 13, 11, 72]. Cell deformation influences local membrane stresses,volume changes, and the cytoskeleton organization, which may be involved in early celldamage [12, 11].

The role of deformation in the development of tissue damage was investigated byseveral researchers. Peeters et al. developed a single cell loading device to monitor thebiomechanical response of skeletal muscle cells under sustained compression [61]. Duringcompression the volume of the cell remained constant, but the surface area of the cellincreased, which was probably caused by stretching of the cell membrane. In a laterstudy, Peeters et al. also found that the deformation of the cells was anisotropic, sincethe cells deformed from an elliptical to a more circular shape [62]. It was suggested thatthis anisotropic deformation was due to the preferred orientation of actin filaments. Withthis loading device, the mechanical and failure properties of attached muscle cells werealso determined [63], as well as the viscoelastic properties of the cells [64]. Bouten et al.developed a three-dimensional in vitro model system, consisting of muscle cells seededand cultured in agarose [12]. Cylindrical constructs cultured up to day 12 were subjectedto 20% gross strain for periods of 1, 2, 4, 12, and 24 h. For all straining periods the

18

Chapter 3. Damage pathways 3.4. Sustained deformation of cells

total percentage of cell damage appeared to be significantly higher in strained constructsthan in unstrained controls. In addition, Breuls et al. developed a compression device tosimultaneously compress tissue engineered muscle constructs with circular, impermeableglass indenters [14]. Results showed that dead cells were highly localized below theindenter and that higher strains led to earlier damage initiation. Furthermore, Gawlittaet al. exposed tissue engineered muscle constructs to different combinations of hypoxiaand deformation, and found that deformation had decremental effects on tissue viability,but that hypoxia had no additional effect within 22 hours [28].

The effect of deformation on tissue damage was also studied in animal models. Bos-boom et al. investigated the location of damage development by applying compressiveloading to the tibialis anterior muscle of rats [10]. They reported that besides histology,also MRI can be used to find locations of damage, since increased signal intensities inT2-weighted MR images correlated with damage determined by histological examina-tion. Stekelenburg et al. also examined muscle damage after compressive loading of rattibialis anterior muscle with T2-weighted MRI in combination with histological exami-nation [78]. In contrast to the study of Bosboom et al. [10], the loading procedure inthis study also took place in the MR scanner, so the exact indentation of the musclecould be observed, see figure 3.5.

Figure 3.5: Transversal T2-weighted-sum images taken before loading (A), during loading (B), andimmediately after unloading (C). The TA muscle is indicated by the white line in A [78].

In the muscle underneath the indenter a necrotic region was observed, which wasassumed to be caused by the large deformation that pulled the muscle fibers apart.It was also noted that T2 values in slices distal to the indenter were higher than inslices proximal to the indenter. This was probably due to a different perfusion status ofthe tissue, since the large indentation presumably led to the collapse of blood vessels,whereas in the tissue proximal to the indentor the perfusion was hardly affected duringloading. The difference in amount of affected volume largely disappeared during thefirst hour after unloading, which indicates that the effect of ischemia was reversiblewithin the muscle tissue. However, it was also noted that ischemia might accelerate orincrease tissue damage. In a later study, Stekelenburg et al. studied muscle damage inrat tibialis anterior muscle during ischemia by applying a cuff around the thigh, andduring ischemia and deformation using an indentor applied to the muscle. This revealedthat ischemia alone only led to reversible damage, whereas the combination of ischemia

19

Chapter 3. Damage pathways 3.4. Sustained deformation of cells

and deformation was far more damaging [79].Breuls et al. developed a multilevel finite element model to investigate local cell

deformations in engineered tissue constructs, subjected to macroscopic loads [16]. Theresults showed that there were large differences in strain energy density within the mi-crostructure at one macroscopic point, indicating that individual cells can experiencedeformations which highly exceed the macroscopic deformation. Therefore, it should bequestioned whether cell damage due to deformation can be evaluated based on macro-scopic deformations. Breuls et al. also used this multilevel finite element approachto model the development of tissue damage in muscle tissue which is compressed overa bony prominence [15]. In this study, a damage law was derived from in vitro ex-periments with tissue engineered skeletal muscle constructs to evaluate tissue damage[14]. It was reported that muscle damage started at locations deep within the muscletissue layer, where shear strains were largest. A parameter study also showed that alower cell stiffness resulted in an increased area of tissue damage and a faster damagegrowth. Furthermore, related to the animal experiments of Stekelenburg et al., Ceelenet al. developed dedicated two-dimensional finite element models of the hind limbs ofrats used in the experiments to investigate the strain state of the muscle. The results ofthese models show a correlation between the location of damage and local strain [19],see figure 3.6.

Figure 3.6: Left and middle: maximum shear strains in finite element models of Ceelen et al. basedon two experiments of Stekelenburg et al.. Locations of damage (x) in MR images are superimposed onthe image. Right: number of damaged and undamaged pixels against maximum shear strain. [19]

20

Chapter 4

Discussion

4.1 Summary

Pressure ulcers are caused by sustained mechanical loading; without a mechanical load-ing no pressure ulcer will develop. External loading of the tissues leads to a nonuniforminternal stress and strain state. This internal mechanical state depends on the magni-tude of the applied loading, type of external loading (pressure, shear, friction), and themechanical and geometrical properties of the tissues. In addition, the time of exposureto mechanical loading also plays an important role in the development of tissue damage.On the other hand, the ability of the tissue to withstand mechanical loading determineswhether a certain loading will lead to the development of an ulcer or not.

Both the mechanical loading and patient susceptibility are influenced by intrinsic andextrinsic factors. Intrinsic factors are related to the individual, e.g. with age, mobility,weight, posture, incontinence and nutritional state. Extrinsic factors are related to theenvironment, e.g. with nursing, the contact surface, nutrition, and the temperature andhumidity of the environment. In clinical practice, these risk factors are used in riskassessment scales to identify patients at risk of developing pressure ulcers. However, thevalue of these scales remains questionable. Moreover, SCI patients are always identifiedas high risk patients, while they may stay free of ulcers for a long time, and suddenlydevelop an ulcer in a very short period. Risk assessment scales can therefore not beused in this group of patients.

The underlying pathways whereby mechanical loading leads to tissue breakdown arehardly understood. At the moment theories involve:

• Localized ischemiaMechanical loading can lead to the collapse of blood vessels, thereby stopping bloodflow and depriving the tissue of its oxygen and nutrients supply. Furthermore,ischemia initiates a series of chemical and pH imbalances. In general, muscletissue appears to be tolerant of ischemia for 4 hours, fat up to 13 hours, and skinup to 24 hours [7].

21

Chapter 4. Discussion 4.2. Focus questions for our research

• Ischemia/reperfusion injuryWhen the blood flow to the tissues is restored after ischemia, oxygen combineswith free radical species generated during ischemia to form reactive oxygen species,that also cause cell damage. It is hypothesized that the reperfusion phase is moredamaging than ischemia alone. For example, when the total extent of ischemiais constant, a larger number of reperfusion events during that period results inincreased tissue damage [65, 83]. Furthermore, gradual reperfusion leads to lesstissue damage when compared to conventional reperfusion [26, 84].

• Impaired lymphatic drainageDuring mechanical loading, the lymphatics can also collapse, which leads to anaccumulation of waste products. The exact contribution of this effect to the onsetof tissue damage is however unclear.

• Sustained deformation of cellsCell deformation influences local membrane stresses, volume changes, and thecytoskeletal organization, which may be involved in cell damage [12, 11]. In fact,it was shown in an animal model that ischemia in combination with deformationof cells was more damaging than ischemia alone [79]. In addition, a computermodel also shows that there is a correlation between local tissue deformation anddamage in the animal model [19].

4.2 Focus questions for our research

At this moment, by combining the results of the animal model of Stekelenburg et al.[78, 79] with dedicated finite element models [19], it is already studied whether thereis a correlation between local tissue deformation and damage. Since the local tissuedeformation is important for the development of tissue damage, it would also be inter-esting to investigate how the internal mechanical state of tissues depends on variationsin mechanical and/or geometrical properties of tissues among individuals. Furthermore,it will be studied how the internal mechanical state is influenced by changes in materialand geometrical properties as a result of disease.

It is also known that a certain mechanical loading will not lead to the same amountof damage in each individual, i.e. some individuals are more to susceptible to developpressure ulcers than others. Therefore, if a certain relation between local tissue defor-mation and damage can be derived from the animal experiments, it is important toinvestigate whether such a relation is the same for each animal or not. If it is similar,then the local tissue deformation probably is the main trigger for the development oftissue damage, and thus the susceptibility of an individual to develop a pressure ulcer ismainly a mechanical issue. Otherwise, also other factors play a role in this process, e.g.lack of tissue repair, bad tissue perfusion, and altered metabolism. It would therefore beinteresting to compare the relations between tissue deformation and damage of subjectswith a similar (patho)physiological state, and also to compare these results with a groupof subjects with a different (patho)physiological state.

22

Chapter 4. Discussion 4.2. Focus questions for our research

The overall goal of this project is to develop a biosensor system on the basis ofbiochemical markers to detect deep tissue injury at an early stage. To achieve thisgoal, it is important to know at which moment it is necessary to adopt preventivemeasures to prevent deep tissue injury, and at which moment it is too late. This isillustrated in figure 4.1. In situation a, the amount of tissue damage D increases untilthe load is removed at time ta. At this moment, there is no significant tissue damageyet (D < Dmin) and the tissue will return to its original status. In situation b, there ismuch more damage (Dmin < D < Dmax) and preventive measures are needed to enablethe damaged tissue to heal again. In situation c, the amount or degree of damage haspassed a critical value (D > Dmax), after which it is expected that tissue damage willcontinue to increase even if the load is removed. Probably, it is not easy to detect tissuedamage by visual inspection before it has reached damage criterion Dmax. A biosensorsystem should therefore be able to detect tissue damage in the range Dmin < D < Dmax.An important question for our research is therefore: what are the amounts or kinds ofskeletal muscle damage corresponding with Dmin and Dmax?

Furthermore, biochemical markers have to be chosen that are released upon skeletalmuscle damage. With respect to these markers, it should be investigated how specificthey are for skeletal muscle damage and also how large the minimal amount of damageis that can be detected with these markers. A theoretical model may also be developedto gain more insight into the kinetics of these biomarkers in injured skeletal muscle.

t

D

Dmax

a

b

c

ta tb tc

too early

too late

take preventivemeasures

Dmin

Figure 4.1: Illustration of how tissue damage is expected to change in time due to mechanical loading.Below a certain damage criterion Dmin, the damage is not considered to be significant. Above damagecriterion Dmax tissue damage is expected to increase even if the load is removed. Between these twocriteria, preventive measures can probably lead to healing of the tissue.

23

Bibliography

[1] http://webpages.charter.net/nursing project/images/progression-of-a-decubitis-ulcer.jpg.

[2] M. A. Ankrom, R. G. Bennett, S. Sprigle, D. Langemo, J. M. Black, D. R. Berlowitz,and C. H. Lyder. Pressure-related deep tissue injury under intact skin and thecurrent pressure ulcer staging systems. Adv. Skin Wound. Care, 18(1):35–42, 2005.

[3] L. Bennett and B. Y. Lee. Pressure versus shear in pressure sore causation. InB. L. Lee, editor, Chronic ulcers of the skin., pages 39–56. McGraw-Hill, 1985.

[4] L. Bennett and B. Y. Lee. Vertical shear existence in animal pressure thresholdexperiments. Decubitus., 1(1):18–24, 1988.

[5] N. Bergstrom. Patients at risk for pressure ulcers and evidence-based care forpressure ulcer prevention. In D. L. Bader, C. V. C. Bouten, D. Colin, and C. W. J.Oomens, editors, Pressure ulcer research; current and future perspectives., pages35–50. Springer-Verlag, 2005.

[6] J. Black, M. Baharestani, J. Cuddigan, B. Dorner, L. Edsberg, D. Langemo, M. E.Posthauer, C. Ratliff, and G. Taler. National pressure ulcer advisory panel’s up-dated pressure ulcer staging system. Urol. Nurs., 27(2):144–50, 156, 2007.

[7] F. W. Blaisdell. The pathophysiology of skeletal muscle ischemia and the reperfu-sion syndrome: a review. Cardiovasc. Surg., 10(6):620–630, 2002.

[8] M. R. Bliss. Aetiology of pressure sores. Rev. Clin. Gerontol., 3:379–397, 1993.

[9] F. M. Boffi, J. Ozaki, N. Matsuki, M. Inaba, E. Desmaras, and K. Ono. Effects ofchemical ischemia on purine nucleotides, free radical generation, lipids peroxidationand intracellular calcium levels in c2c12 myotube derived from mouse myocytes. J.Vet. Med. Sci., 64(6):483–488, 2002.

[10] E. M. Bosboom, C. V. Bouten, C. W. Oomens, F. P. Baaijens, and K. Nicolay.Quantifying pressure sore-related muscle damage using high-resolution mri. J. Appl.Physiol, 95(6):2235–2240, 2003.

[11] C. V. Bouten, R. G. Breuls, E. A. Peeters, C. W. Oomens, and F. P. Baaijens. Invitro models to study compressive strain-induced muscle cell damage. Biorheology,40(1-3):383–388, 2003.

24

Bibliography Bibliography

[12] C. V. Bouten, M. M. Knight, D. A. Lee, and D. L. Bader. Compressive deformationand damage of muscle cell subpopulations in a model system. Ann. Biomed. Eng,29(2):153–163, 2001.

[13] C. V. Bouten, C. W. Oomens, F. P. Baaijens, and D. L. Bader. The etiology ofpressure ulcers: skin deep or muscle bound? Arch. Phys. Med. Rehabil., 84(4):616–619, 2003.

[14] R. G. Breuls, C. V. Bouten, C. W. Oomens, D. L. Bader, and F. P. Baaijens.Compression induced cell damage in engineered muscle tissue: an in vitro model tostudy pressure ulcer aetiology. Ann. Biomed. Eng, 31(11):1357–1364, 2003.

[15] R. G. Breuls, C. V. Bouten, C. W. Oomens, D. L. Bader, and F. P. Baaijens. Atheoretical analysis of damage evolution in skeletal muscle tissue with reference topressure ulcer development. J. Biomech. Eng, 125(6):902–909, 2003.

[16] R. G. Breuls, B. G. Sengers, C. W. Oomens, C. V. Bouten, and F. P. Baaijens.Predicting local cell deformations in engineered tissue constructs: a multilevel finiteelement approach. J. Biomech. Eng, 124(2):198–207, 2002.

[17] R. S. Candadai and N. P. Reddy. Stress distribution in a physical buttock model:effect of simulated bone geometry. J. Biomech., 25(12):1403–1411, 1992.

[18] K. K. Ceelen, C. W. Oomens, and F. P. Baaijens. Microstructural analysisof deformation-induced hypoxic damage in skeletal muscle. Biomech. Model.Mechanobiol., 2007.

[19] K.K. Ceelen, 2007.

[20] W. W. Chow and E. I. Odell. Deformations and stresses in soft body tissues of asitting person. J. Biomech. Eng., 100:79–87, 1978.

[21] R. K. Daniel, D. L. Priest, and D. C. Wheatley. Etiologic factors in pressure sores:an experimental model. Arch. Phys. Med. Rehabil., 62(10):492–498, 1981.

[22] R. K. Daniel, D. Wheatley, and D. Priest. Pressure sores and paraplegia: anexperimental model. Ann. Plast. Surg., 15(1):41–49, 1985.

[23] T. Defloor, L. Schoonhoven, V. Katrien, J. Weststrate, and D. Myny. Reliabilityof the european pressure ulcer advisory panel classification system. J. Adv. Nurs.,54(2):189–198, 2006.

[24] S. M. Dinsdale. Decubitus ulcers: role of pressure and friction in causation. Arch.Phys. Med. Rehabil., 55(4):147–152, 1974.

[25] J. Donnelly. Should we include deep tissue injury in pressure ulcer staging systems?the npuap debate. J. Wound. Care, 14(5):207–210, 2005.

25

Bibliography Bibliography

[26] N. K. Durrani, R. Yavuzer, V. Mittal, M. M. Bradford, C. Lobocki, and B. Silber-berg. The effect of gradually increased blood flow on ischemia-reperfusion injury inrat kidney. Am. J. Surg., 191(3):334–337, 2006.

[27] S. L. Garber and D. H. Rintala. Pressure ulcers in veterans with spinal cord injury:a retrospective study. J. Rehabil. Res. Dev., 40(5):433–441, 2003.

[28] D. Gawlitta, W. Li, C. W. Oomens, F. P. Baaijens, D. L. Bader, and C. V. Bouten.The relative contributions of compression and hypoxia to development of muscletissue damage: an in vitro study. Ann. Biomed. Eng, 35(2):273–284, 2007.

[29] D. Gawlitta, C. W. Oomens, D. L. Bader, F. P. Baaijens, and C. V. Bouten.Temporal differences in the influence of ischemic factors and deformation on themetabolism of engineered skeletal muscle. J. Appl. Physiol, 103(2):464–473, 2007.

[30] A. Gefen. Risk factors for a pressure-related deep tissue injury: a theoretical model.Med. Biol. Eng Comput., 45(6):563–573, 2007.

[31] A. Gefen, N. Gefen, E. Linder-Ganz, and S. S. Margulies. In vivo muscle stiff-ening under bone compression promotes deep pressure sores. J. Biomech. Eng,127(3):512–524, 2005.

[32] P. C. Grisotto, A. C. dos Santos, J. Coutinho-Netto, J. Cherri, and C. E. Piccinato.Indicators of oxidative injury and alterations of the cell membrane in the skeletalmuscle of rats submitted to ischemia and reperfusion. J. Surg. Res., 92(1):1–6,2000.

[33] K. E. Groth. Klinische Beobachtungen und experimentelle studien ber die Entste-hung des Dekubitus. 1942.

[34] A. C. Guyton and J. E. Hall. Textbook of Medical Physiology. 9th edition, 1996.

[35] D. A. Hobson. Comparative effects of posture on pressure and shear at the body-seat interface. J. Rehabil. Res. Dev., 29(4):21–31, 1992.

[36] R. Houwing, M. Overgoor, M. Kon, G. Jansen, B. S. van Asbeck, and J. R. Haal-boom. Pressure-induced skin lesions in pigs: reperfusion injury and the effects ofvitamin e. J. Wound. Care, 9(1):36–40, 2000.

[37] T. Husain. An experimental study of some pressure effects on tissues, with referenceto the bed-sore problem. J. Pathol. Bacteriol., 66(2):347–358, 1953.

[38] K. Ikebe, T. Kato, M. Yamaga, J. Hirose, T. Tsuchida, and K. Takagi. Increasedischemia-reperfusion blood flow impairs the skeletal muscle contractile function. J.Surg. Res., 99(1):1–6, 2001.

[39] M. Kosiak. Etiology and pathology of ischemic ulcers. Arch. Phys. Med. Rehabil.,40(2):62–69, 1959.

26

Bibliography Bibliography

[40] M. Kosiak. Etiology of decubitus ulcers. Arch. Phys. Med. Rehabil., 42:19–29, 1961.

[41] T. A. Krouskop. A synthesis of the factors that contribute to pressure sore forma-tion. Med. Hypotheses, 11(2):255–267, 1983.

[42] T. A. Krouskop, N. P. Reddy, W. A. Spencer, and J. W. Secor. Mechanisms ofdecubitus ulcer formation–an hypothesis. Med. Hypotheses, 4(1):37–39, 1978.

[43] E. M. Landis. Micro-injection studies of capillary blood pressure in human skin.Heart, 15:209–228, 1930.

[44] E. Linder-Ganz, S. Engelberg, M. Scheinowitz, and A. Gefen. Pressure-time celldeath threshold for albino rat skeletal muscles as related to pressure sore biome-chanics. J. Biomech., 39(14):2725–2732, 2006.

[45] E. Linder-Ganz and A. Gefen. Mechanical compression-induced pressure sores in rathindlimb: muscle stiffness, histology, and computational models. J. Appl. Physiol,96(6):2034–2049, 2004.

[46] E. Linder-Ganz, N. Shabshin, Y. Itzchak, and A. Gefen. Assessment of mechanicalconditions in sub-dermal tissues during sitting: a combined experimental-mri andfinite element approach. J. Biomech., 40(7):1443–1454, 2007.

[47] LPZ-projectgroep, 2007.

[48] A. F. Mak, L. Huang, and Q. Wang. A biphasic poroelastic analysis of the flowdependent subcutaneous tissue pressure and compaction due to epidermal loadings:issues in pressure sore. J. Biomech. Eng, 116(4):421–429, 1994.

[49] J. Maklebust. Policy implications of using reverse staging to monitor pressure ulcerstatus. Adv. Wound. Care, 10(5):32–35, 1997.

[50] W. O. McKinley, A. B. Jackson, D. D. Cardenas, and M. J. DeVivo. Long-termmedical complications after traumatic spinal cord injury: a regional model systemsanalysis. Arch. Phys. Med. Rehabil., 80(11):1402–1410, 1999.

[51] C. C. Michel and H. Gillott. Microvascular mechanisms in stasis and ischaemia. InD. L. Bader, editor, Pressure sores: clinical practice and scientific approach., pages153–163. McMillan Press, 1990.

[52] G. E. Miller and J. Seale. Lymphatic clearance during compressive loading. Lym-phology, 14(4):161–166, 1981.

[53] O. W. Neumark. Deformation, not pressure, is the prime cause of pressure sores.Care. Sci. Pract., 1:41–43, 1981.

[54] G. T. Nola and L. M. Vistnes. Differential response of skin and muscle in theexperimental production of pressure sores. Plast. Reconstr. Surg., 66(5):728–733,1980.

27

Bibliography Bibliography

[55] Health Council of the Netherlands. Health council of the netherlands: pressureulcers. Report, 1999.

[56] C. Oomens. Perspectives of numerical modelling in pressure ulcer research. InD. L. Bader, C. V. C. Bouten, D. Colin, D. Colin, and C. W. J. Oomens, editors,Pressure ulcer research; current and future perspectives., pages 149–159. Springer-Verlag, 2005.

[57] C. W. Oomens, O. F. Bressers, E. M. Bosboom, C. V. Bouten, and D. L. Blader.Can loaded interface characteristics influence strain distributions in muscle adjacentto bony prominences? Comput. Methods Biomech. Biomed. Engin., 6(3):171–180,2003.

[58] National Pressure Ulcer Advisory Panel. Pressure ulcers prevalence, cost and riskassessment: consensus development conference statement. Decubitus., 2(2):24–28,1989.

[59] National Pressure Ulcer Advisory Panel. Position on reverse staging of pressureulcers. Adv. Wound. Care, 8(6):32–33, 1995.

[60] L. C. Parish, P. Lowthian, and J. A. Witkowski. The decubitus ulcer: many ques-tions but few definitive answers. Clin. Dermatol., 25(1):101–108, 2007.

[61] E. A. Peeters, C. V. Bouten, C. W. Oomens, and F. P. Baaijens. Monitoring thebiomechanical response of individual cells under compression: a new compressiondevice. Med. Biol. Eng Comput., 41(4):498–503, 2003.

[62] E. A. Peeters, C. V. Bouten, C. W. Oomens, D. L. Bader, L. H. Snoeckx, and F. P.Baaijens. Anisotropic, three-dimensional deformation of single attached cells undercompression. Ann. Biomed. Eng, 32(10):1443–1452, 2004.

[63] E. A. Peeters, C. W. Oomens, C. V. Bouten, D. L. Bader, and F. P. Baaijens.Mechanical and failure properties of single attached cells under compression. J.Biomech., 38(8):1685–1693, 2005.

[64] E. A. Peeters, C. W. Oomens, C. V. Bouten, D. L. Bader, and F. P. Baaijens.Viscoelastic properties of single attached cells under compression. J. Biomech.Eng, 127(2):237–243, 2005.

[65] S. M. Peirce, T. C. Skalak, and G. T. Rodeheaver. Ischemia-reperfusion injury inchronic pressure ulcer formation: a skin model in the rat. Wound. Repair Regen.,8(1):68–76, 2000.

[66] N. P. Reddy and G. V. Cochran. Interstitial fluid flow as a factor in decubitus ulcerformation. J. Biomech., 14(12):879–881, 1981.

[67] N. P. Reddy, H. Patel, G. V. Cochran, and J. B. Brunski. Model experimentsto study the stress distributions in a seated buttock. J. Biomech., 15(7):493–504,1982.

28

Bibliography Bibliography

[68] N. P. Reddy and K. Patel. A mathematical model of flow through the terminallymphatics. Med. Eng Phys., 17(2):134–140, 1995.

[69] S. M. Reichel. Shearing force as a factor in decubitus ulcers in paraplegics. J. Am.Med. Assoc., 166(7):762–763, 1958.

[70] J. B. Reswick and J. E. Rogers. Experience at rancho los amigos hospital withdevices and techniques that prevent pressure sores. In R. M. Kenedi and J. M.Cowden, editors, Bedsore biomechanics, pages 301–310. The Macmillan Press, 1stedition, 1976.

[71] E. Rubin, F. Gorstein, R. Rubin, R. Schwarting, and D. Strayer. Rubin’s pathol-ogy: clinicopathologic foundations of medicine. Lippincott Williams & Wilkins, 4thedition, 2005.

[72] T. J. Ryan. Cellular responses to tissue distortion. In D. L. Bader, editor, Pressuresores: clinical practice and scientific approach., pages 141–152. McMillan Press,1990.

[73] A. H. Sacks. Theoretical prediction of a time-at-pressure curve for avoiding pressuresores. J. Rehabil. Res. Dev., 26(3):27–34, 1989.

[74] R. Salcido, J. C. Donofrio, S. B. Fisher, E. K. LeGrand, K. Dickey, J. M. Carney,R. Schosser, and R. Liang. Histopathology of pressure ulcers as a result of sequentialcomputer-controlled pressure sessions in a fuzzy rat model. Adv. Wound. Care,7(5):23–4, 26, 28, 1994.

[75] L. Schoonhoven, J. R. Haalboom, M. T. Bousema, A. Algra, D. E. Grobbee, M. H.Grypdonck, and E. Buskens. Prospective cohort study of routine use of risk assess-ment scales for prediction of pressure ulcers. BMJ, 325(7368):797, 2002.

[76] A. Smith, G. Hayes, A. Romaschin, and P. Walker. The role of extracellular calciumin ischemia/reperfusion injury in skeletal muscle. J. Surg. Res., 49(2):153–156, 1990.

[77] A. Stekelenburg, C. Oomens, and D. Bader. Compression-induced tissue damage:animal models. In D. L. Bader, C. V. C. Bouten, D. Colin, D. Colin, and C. W. J.Oomens, editors, Pressure ulcer research; current and future perspectives., pages187–204. Springer-Verlag, 2005.

[78] A. Stekelenburg, C. W. Oomens, G. J. Strijkers, K. Nicolay, and D. L. Bader.Compression-induced deep tissue injury examined with magnetic resonance imagingand histology. J. Appl. Physiol, 100(6):1946–1954, 2006.

[79] A. Stekelenburg, G. J. Strijkers, H. Parusel, D. L. Bader, K. Nicolay, and C. W.Oomens. Role of ischemia and deformation in the onset of compression-induced deeptissue injury: Mri-based studies in a rat model. J. Appl. Physiol, 102(5):2002–2011,2007.

29

Bibliography Bibliography

[80] Q. Sun, F. Lin, S. Al-Saeede, L. Ruberte, E. Nam, R. Hendrix, and M. Makhsous.Finite element modeling of human buttock-thigh tissue in a seated posture. SummerBioengineering Conference, 2005.

[81] D. Thompson. A critical review of the literature on pressure ulcer aetiology. J.Wound. Care, 14(2):87–90, 2005.

[82] B. A. Todd and J. G. Thacker. Three-dimensional computer model of the humanbuttocks, in vivo. J. Rehabil. Res. Dev., 31(2):111–119, 1994.

[83] S. Tsuji, S. Ichioka, N. Sekiya, and T. Nakatsuka. Analysis of ischemia-reperfusioninjury in a microcirculatory model of pressure ulcers. Wound. Repair Regen.,13(2):209–215, 2005.

[84] S. Unal, S. Ozmen, Y. DemIr, R. Yavuzer, O. LatIfoglu, K. Atabay, and M. Oguz.The effect of gradually increased blood flow on ischemia-reperfusion injury. Ann.Plast. Surg., 47(4):412–416, 2001.

[85] J. R. Urbaniak, A. V. Seaber, and L. E. Chen. Assessment of ischemia and reper-fusion injury. Clin. Orthop. Relat Res., (334):30–36, 1997.

[86] J. D. Zhang, A. F. Mak, and L. D. Huang. A large deformation biomechanicalmodel for pressure ulcers. J. Biomech. Eng., 119(4):406–408, 1997.

30