Events in the tissues

The first events begin on the surface of the skin, when it is touched. It is now widely accepted that human touch is a prerequisite for the healthy functioning of the individual. At the same time, in the area of health care, prolific technological advances are decreasing our opportunities for physical intervention. In addition, Western culture is largely moving away from physical activity towards seated activities, such as those involving computers or performing arts entertainment with the result that many are losing the ability to integrate body and mind, suffering consequently from a lack of sensory unity. Touching and being touched is a basic human need but there is little opportunity outside of families to express or receive feelings of care by touching. This situation is magnified for many elderly people and those residing in an institutional care environment. Massage can offer valid human contact to counterbalance the potentially dehumanising effects of tactile deprivation.

The anatomy and physiology of touch receptors and their interaction with the central nervous system (CNS) were discussed in Chapter 2. We know and understand a great deal about the mechanisms of touch; what is less clear is how the physical phenomenon of touch affects our moods, emotions and levels of autonomic, cortical and behavioural arousal. Knowledge of this would enable us to predict the responses that might occur as a result of any tactile intervention.

The sense of touch is one of the earliest senses to develop. The human embryo has been observed to withdraw reflexly from stroking stimuli at 6 weeks after conception (Montagu 1978). Preterm infants have been observed not to tolerate massage but to prefer ‘containment’, that is, holding or cuddling, which is thought not to stimulate developmentally uninhibited reflexes (Hartelius et al 1992). Containment may be perceived by the infant as a similar stimulus to the pressure exerted by amniotic fluid in the uterus. However, a later study (Field et al 1986) found that not only did preterm infants tolerate massage, but they showed increased weight, alertness and maturity. Further work by this team (Diego et al 2007) has shown that massaging pre-term infants results in short-term increased vagal activity and gastric motility which could explain the increased weight gain. Mendes and Procianoy (2008) found that massage given by mothers four times each day reduced length of hospital stay. The babies were, however, also given passive limb movements. An earlier systematic review by Vickers et al (2001) of studies in larger babies had shown that work conducted before this on pre-term and low birth-weight infants was inconclusive as the positive effects were small and therefore of low clinical significance.

Observations on young animals have provided some information about their response to touch, the implications of which may be transferable to humans. Among many mammals tactile rituals occur after birth which serve physiological functions and may promote normal emotional development. There is a relationship between the stroking of young animals by humans and a reduction in the animal’s physiological response to stress, demonstrated by a decreased output of adrenocorticotrophic hormone (ACTH) (Seyle 1950) and reduction in blood pressure and heart rate (Lund et al 1999). Young animals that are handled also show greater development of the cortex and subcortex of the brain; they learn faster and have a more advanced stage of neural development than non-handled animals (Ruegammer et al 1954). Resistance to infection later in life may also be influenced beneficially by cutaneous stimulation experienced by the infant animal (Soloman & Moos 1964).

It is thus probable that cutaneous stimulation during critical periods of a human infant’s development will promote similar normal organic and behavioural growth. After the age of about 3 months, the infant begins to utilise his/her sense of touch to learn about the immediate environment. Deprivation of this primary learning process may compromise later social and learning responses, and adversely affect the way the adult utilises tactile information (Frank 1957, Mason 1985).

Touch

Throughout our lives, contact with the environment ensures that our skin receives continual stimulation. We receive tactile messages from, for example, the clothes we wear, from the water we wash in, from the rubbing of lotions on our skin, and from people and animals we touch or are touched by. Depending on the quality of tactile stimulation perceived, we can feel such emotions as pleasure, pain, fear or revulsion. Similar feelings are engendered by the type of human contact we experience, which is often tempered by integration with other somatic senses. Contrast, for example, the feelings experienced by the touch of a warm hand or a cold one; the sense of smell may influence our response; auditory and visual information may also influence our response to touch. Usually we permit touch when we perceive it as safe or desirable but avoid physical contact with people whom we dislike. The subjective experience of the touch determines our neural and hormonal responses, which in turn influence our integrated response to the touch. As massage necessarily involves touch, the therapist needs to consider not only the mechanical and physiological effects but also the emotional responses likely to result. This careful thought will begin before the massage, while taking the client’s medical history and during the assessment.

Therapeutic considerations

The first fact to be established is whether the client has given permission to be touched. A major factor will be how the patient perceives the therapist and the environment. The variables may include the way the therapist presents herself, the degree of privacy involved, and whether the environment is perceived to be safe. Careful attention should be given to differentiating between verbal permission and any non-verbal indicators to the contrary. Some individuals involuntarily interpret touch as threatening, and the sympathetic nervous system may be aroused as a result; these individuals are described as being ‘tactually defensive’.

An explanation should be given of the method of massage to be used; a patient who has previously experienced only a pleasant skin rub with essential oils may be unprepared for deep tissue manipulations. The quality of touch which is conveyed to the patient may be predetermined by the therapist’s intention and may vary through caring, sensitive and professionally based to cold, hasty and clumsy. Invasion of the body surface by unwelcome types of touch will be perceived as threatening or undesirable, and is likely to produce an unwanted effect. Conversely, the effects of appropriate touch will enhance the rapport between patient and therapist, thus enabling further therapeutic intervention. Constant monitoring of the patient is required throughout treatment to ensure that changes in technique are made when appropriate.

As professional therapists, we are concerned with all the effects our massage may produce. When administering a massage the main objective is often to have a mechanical effect with the aim of restoring normal function. However, we should not neglect the fact that there may be other stimuli, extrinsic or intrinsic to the massage, which are capable of producing undesirable effects. Such factors as the environment, the temperature of the environment, the degree of privacy and background noise will all affect the response of the client; similarly, the various sensations provoked by the quality of touch can produce a variety of responses. Our responsibility as therapists is to ensure that all the effects of our treatment are desirable and none detract from our aims. In this respect the therapist should have an understanding of autonomic and emotional arousal.

Stressors

The main centres of ANS activation are in the spinal cord, brainstem and hypothalamus, with control also being influenced by the limbic cortex. Visceral functions such as arterial pressure, heart rate, gastrointestinal motility and secretion, temperature and sweating are controlled. When the system is working optimally, there is a state of homoeostatic equilibrium. The ANS is also influenced through visceral reflexes, sensory signals that trigger reflex responses of the visceral organs. Change of function can be rapid: the heart rate can increase to twice its normal level in 3–5 seconds, and arterial pressure can double in under 15 seconds or conversely, arterial pressure may increase to 50% of normal in 10 to 40 seconds (Guyton & Hall, 2006). (See Thibodeau et al (2007), pp 842–3, Figs. 22.2–22.4.)

Any agent that provokes sympathetic arousal is termed a ‘stressor’. Stressors may be physical, psychological or sociocultural (Seyle 1982). One current theory of stress is that it is cognitively controlled; that is, an individual’s response to a stressor is dependent upon that particular individual’s previous experience of similar stessors, and his/her present ability to cope with the stressor. While a moderate level of sympathetic arousal is desirable to facilitate most everyday activities—it keeps us mentally and physically alert—prolonged exposure to a stressor that produces a high level of autonomic arousal can have undesirable physiological effects, resulting in, for example, decreased immunity, and increases in hypertension and vascular disorders (Willard 1995).

Current concepts of stress take account of the neuroendocrine changes in the body in response to a stimulus (see Fig. 10.1, which summarises some of the current concepts of stress responses).

An emotion is an expression of subjective feeling accompanied by neural and hormonal activity; emotions are determined by learned, cognitive and biological factors. The systems that control autonomic and emotional activity are interactive, being linked by neural impulses and hormones. A therapist who has a good understanding of the integration between these systems will work holistically with clients, thus ensuring that the whole person benefits from her intervention.

The tissue layers

Changes can actually be seen on the surface of the skin during vigorous massage, when some reddening occurs. The amount depends to some extent on the reactivity of the skin, which is determined by skin type, although vasodilatory reactions in the skin are common. Fundamentally, it is as a result of release of histamine from the mast cells. Mast cells are found in connective tissue and contain histamine, heparin and hyaluronic acid and it is known that cells respond to mechanical signals (Banes et al 1995). Cell deformation may activate calcium ion channels and influence calcium transport. Mechanical stress has specifically been found to activate mast cell secretion (Theoharides 1996). It is unclear why this mechanical irritation and its resulting vasodilatation should occur, and to what purpose. Further reddening seen in the skin may be due to shear forces acting on the endothelium of blood vessels, causing release of the vasodilator nitric oxide (Noris et al 1995), which is angioprotective. Vasodilatation is accompanied by an increase in capillary permeability and it is likely that the tissue fluid released from these capillaries has a flushing effect on the tissues, both removing irritants and allowing protective chemicals to be brought to the area via the bloodstream. The reddening may increase if the hands glide over the surface of the skin, particularly with speed as this increases friction. Reddening can be reduced by massaging more slowly and by using an oily medium.

When a hand is held over the surface of another person’s skin, heat can be felt. If the hand is placed on skin and held in a stationary position, this heat can be felt to increase. Rubbing over the surface of the skin causes friction and this increases heat even further. Heat is a form of energy and some schools of massage, particularly those grounded in Eastern practices, base their approach on a belief that an energy field exists around the body. There is, however, no evidence for this aura, as it is known, and the Kirlian photography used to measure it has been found to be scientifically invalid. Some therapies are based on a belief in energy fields running through the body along specific pathways known as meridians. These principles are utilised in orthodox health care systems through acupuncture (mainly used by physiotherapists and doctors), reflex therapy (used by physiotherapists) and therapeutic touch (popular with nurses in the USA). Therapeutic touch, non-contact therapeutic touch and other energetic forms of massage use this principle to varying degrees as they attempt to normalise the energy fields, promoting healing and well being. Within biomedicine, it is often interpreted as ANS activity, as discussed below.

With a slight increase in pressure, layers of tissue are moved with the hands, rather than the hands gliding lightly over the skin surface. A very light glide necessitates some movement of the epidermis. If there is friction between the therapist’s hand and the patient’s skin, the epidermis moves with the therapist’s hands and is gently stretched. As this layer is so thin, the dermis must move simultaneously because of traction between the dermis and epidermis which results from natural adherence between the layers. The application of slightly more pressure with friction (but note the massage still feels very light) and a traction effect occurs between the dermis and subcutis. Resistance or tension is felt when this traction reaches its limit and all layers are stretched. This is referred to as the end-feel of the stroke and at this point, if the hands glide or continue to push into the resistance, the massage is deepened as traction occurs at the next interface down.

So far, we have recognised that massage involves an interaction of energy between the patient and therapist; that it utilises the effects of touch to induce relaxation, communication and a sense of well being; and that it produces movement of the tissues in subsequent layers as a result of traction at tissue interfaces. In addition, a complete variety of strokes lifts, pulls, squeezes and twists the skin, connective tissue, tendons, ligaments, muscles, blood vessels and nerves. Sensory and autonomic nerves are stimulated, inducing changes in the nervous and circulatory systems, and movement is effected in abnormal tissue, for example scar tissue or where layers are adherent.

The therapist should be aware that whilst it is possible to target a specific structure, the surrounding tissues will also be influenced due to their interconnectedness, particularly those in superficial layers.

Circulatory effects

The pressure of the massage itself increases pressure in the tissues. Pressure gradients are created between the tissue spaces and vessels as discussed in Chapter 2. As the hands are moved, so the area of increased pressure is moved, creating a fluctuating pressure difference between one area of tissue and another. Thus fluid moves constantly from tissues to vessels and back again, as it flows from areas of high to low pressure. This can occur in two ways: if pressure is increased only in the tissue space and not in the vessel, fluid will move from the tissue into the vessel. Slightly more pressure, however, also increases the pressure in the vessel, so there may be a tendency for fluid to move out of the vessel into tissue space which is at lower pressure. If the vessel is compressed and this pressure moves longitudinally along its course, as in effleurage, then the fluid is pushed proximally along the vessel, leaving the collapsed vessel behind the hand to fill again rapidly. This refilling or milking effect can push fluid towards the heart. If it occurs in veins, a suction-like effect will take place, aided by the valves which prevent back-flow. In addition, manipulation of the tissues at a careful depth will cause a pull on the filaments, which are connected to the flaps in the walls of lymphatic vessels, allowing larger plasma proteins to be removed from the tissue spaces, thus restoring a normal osmotic pressure to the extracellular fluid.

While there appear to be several ways in which fluid balance in the tissues is influenced, the mechanisms are probably more complex than the theoretical supposition described here. It is not possible for the therapist to know exactly where the smaller vessels lie as their exact positions are subject to individual variation. The deep vessels cannot be palpated if they are normal and healthy. The massage manipulations themselves are complex and involve a combination of squeezing, stretching, pulling and traction forces, with movement occurring in different directions and in different tissue planes: for example, kneading consists of circular movements, skin rolling produces transverse movements, stroking and effleurage produce longitudinal forces. Consequently, there are complex repetitive pressure changes occurring in varying directions and at different depths. This is likely to have an effect on fluid interchange, whereby fluid is pushed from the tissue spaces into the vessel, towards the lymph nodes and heart, and new fluid is pushed or drawn into the spaces. Generally, it seems less logical to assume that massage will reduce the amount of fluid in the spaces when it is more likely to replenish it.

This flushing effect in the tissues is important. New circulation is brought to the area, bringing with it fresh nutrition, and the stasis by which inflammatory products, chemical irritants and toxins linger in the tissues is corrected. The local environment is therefore changed for the better. The mechanism by which chemical irritants in the tissues can cause an undesirable plasticity in the spinal cord, lowering the threshold to pain within a whole neuronal pool, is discussed further in Chapter 4. Replenishing tissue fluid and removing inflammatory products will reduce this effect, possibly preventing or reducing some types of chronic pain. Removing metabolites and chemicals such as potassium from muscles by releasing them from muscle cells and ‘flushing’ the muscle tissue with new circulation will reduce muscle soreness following exercise. This effect will reduce pain in situations where metabolites have built up due to prolonged muscle spasm, increased tone (for example, where there is excessive anxiety or tension) and conditions such as fibromyalgia. Indirectly, massage can promote healing by bringing new circulation to the area. Occasionally, in a chronically swollen limb or when fluid is trapped in a tissue space (as in the hand or around the ankle, for example), or where there is fibrous swelling, massage can soften and release the swelling, facilitating its removal.

A much-cited study into massage and blood flow was carried out by Wakim et al in 1949. This team measured blood flow by plethysmography and spirometry in the forearms and hands and in the lower legs. In a group of 15 asymptomatic subjects they found that arm massage increased blood flow to significant levels in 11 of 12 observations and that leg massage increased it to significant levels in 11 of 14 observations. This increase was maintained 30 minutes after the cessation of the massage. The massage given in this instance was a vigorous type involving deep stroking, deep forceful kneading and friction. After a modified Hoffa type of massage, which included stroking and deep kneading, 16 of 32 readings on the upper limbs showed insignificant results while 10 of 32 showed an increase and 6 of 32 a decrease in blood flow. Lower limb readings were insignificant in 15 of 24 subjects, with 5 of 24 showing an increase and 4 of 24 a decrease. In paralysed limbs, 4 of 6 had a significant increase and 3 of 5 limbs with spastic paralysis showed an increase. Rheumatoid limbs showed readings fairly evenly distributed across the significant increase, non-significant increase and reduced flow categories. Of further interest is the fact that, following two sessions in two patients with poliomyelopathy, all observations were significant. This research was conducted some time ago and plethysmography has been criticised by Hansen and Kristensen (1973) as a technique which only measures blood flow in the skin. Significance levels were arbitrarily set at 15% and improvement was measured in percentages. There was no statistical analysis, so it is not a statistical significance referred to here. There was no account taken of probability in the calculations; the results are of clinical significance only if the pattern of results occurs more frequently than they would by chance, which is why statistical analysis is important. The stimulating massage that produced better results is unlikely to be carried out frequently in a real clinical situation, and the trauma and irritation of such a massage would be expected to increase blood flow in skin, but with a limited therapeutic value. The results of this study are of interest but should be treated with caution, as little convincing scientific evidence is offered.

A further attempt to measure the effects of massage on blood flow was made by Hansen and Kristensen (1973) by the more sophisticated ¹³³Xe clearance technique. This substance was injected into the calf muscles and measured with scintillation detectors during and after massage. The ‘centripetal effleurage’ was conducted for 5 minutes in the legs of two women and 10 men, aged 20–32 years. The mean disappearance rate during massage was statistically significant, with the significance reducing 0–2 minutes after the massage. The results show a slight effect, with none in avascular subcutaneous tissues. Unfortunately, little detail was given concerning the massage itself, for example the amount of pressure used. Also, some traditional principles of circulatory massage were not respected, namely, massaging proximal areas before distal ones and using an assortment of techniques. In addition, differences in amounts of massage time were not monitored. This research supports the assertion that massage increases blood flow in muscles but missed the opportunity to increase its impact by informing us more about types of massage and lengths of treatment. The study was, in fact, designed to compare the effects of massage with those of shortwave diathermy and ultrasound (found to be not significant) and therefore did not examine massage in particular detail. As the blood flow effects were short-lived, this would suggest that these effects are as a result of mechanical stimulation of blood vessels rather than chemical or reflex responses.

The following year produced a publication which to some extent substantiated these findings (Hovind & Nielsen 1974). Blood flow was measured in the brachioradialis and vastus lateralis muscles of nine volunteers aged 22–32 years, following intramuscular injection of saline and ¹³³Xe. The results showed that, following tapotement, blood flow rose significantly for 10 minutes after cessation of the technique, but petrissage results were inconsistent, short-lasting and not statistically significant. As the tapotement results were similar to those found during active isometric muscle contractions, it is probable that this technique (described by the authors as ‘unpleasant to the experimental volunteers’) stimulated reflex muscle contractions, which increased the blood flow.

Shoemaker et al (1997) studied the effect of massage to the forearm flexor muscle and the quadriceps on blood flow (mean blood velocity (MBV)). Blood flow was measured by pulsed Doppler ultrasound and vessel diameter by echo Doppler ultrasound. Ten subjects were studied, with readings of MBV taken prior to treatment and at 5, 10, 20 seconds and 5 minutes following the onset of massage. Vessel diameter readings were taken before and after massage. Massage did not significantly increase blood flow in either muscle group, whereas light exercise did elevate blood flow from rest. This is a sound study in which the massage was conducted by a registered massage therapist. It is not indicated whether the massage was superficial (with the hands gliding over the skin, using an oily medium) or whether it was deep. It is logical to assume the effects on blood flow of deep and superficial massage would differ and a comparison of different depths would make an interesting study.

In normal limbs, then, there appears to be some increase in blood flow, either in the skin or intramuscularly, during massage, particularly effleurage, although this is not universally substantiated by all studies. Small sample sizes and inconsistent methodologies leave results inconclusive. Increases in circulation occurring during, but not following, massage suggest that it results from the mechanical, rather than the reflex or chemical effects of massage, although it has been asserted for many years (Carrier 1922) that vasodilatation during fairly vigorous massage occurs as a result of stimulating the axon reflex, observed as a reddening of the skin. Massage has, in fact, been shown to increase the effect of a vasodilator substance in the skin following superficial massage and in muscles following deep massage (Severini & Venerando 1967). As physiological compensating mechanisms are extremely efficient in healthy tissues, any alteration in local blood flow will be compensated for by autoregulatory processes; these researchers are perhaps limiting their results by using subjects with normal circulatory systems. Severini and Venerando also found, surprisingly, that deep and superficial massage decreased skin temperature. Deep massage demonstrated ‘appreciable’ increases in blood flow in both the massaged and non-massaged legs; increases in cardiac stroke volume; and decreases in heart rate and systolic and diastolic arterial pressure. Unfortunately, only the abstract is available in the English language.

Morhenn (2000) found that massaging the cheeks of the face increased skin temperature in seven out of eight human subjects, which plateaued after 40 minutes of massage. It was accompanied by erythema. The effect was blocked by pre-treatment of capsaicin, a chemical which causes release of substance P by peripheral nerve endings which suggests that the raised temperature effect can be partly controlled by substance P. A point of less relevance than interest is that the researchers conclude that social grooming in animals may be necessary because of the survival effect of neurotransmitter release. Zoologists, however, have conducted much research into this subject and their explanations for animal grooming vary from stress reduction, care of the coat/feathers, parasite control, communication and the spread of chemicals (such as pheromones in bees), depending on the species.

Cambron et al (2006) missed an opportunity to measure changes in blood pressure following massage. They studied the effects of massage on 150 clients, but utilised different types of massage for different clients. Unsurprisingly, the painful treatments caused a rise in blood pressure whilst some clients undergoing other forms of massage experienced a drop in blood pressure. The findings of this study should be interpreted with caution.

In 1988, Flowers compared massage with string wrapping and a combination of the two in 56 upper limb digits in 10 women and 4 men aged 24–61 years. They used a retrograde ‘milking’ massage along the whole length of the digit and monitored its effectiveness by measuring the distal interphalangeal joint girth with a tape measure. Using sound statistical tests, they found that neither the string wrap nor the massage alone demonstrated significant results but, when combined, results were significant. Continuous stroking was better than intermittent stroking. This is interesting, as the combined effect would operate at a much deeper level in the tissues than manual lymphatic drainage, which is the current treatment of choice for protein-rich oedema. Presumably, the swelling in this study was more acute and was possibly trapped in the tissue spaces, needing mechanical assistance to return back into the bloodstream. The string wrapping maximised the principle of pressure gradient-induced fluid dynamics, and the massage mechanically aided the process. It is illuminating that a combination of these two modalities produced good results. This indicates that the pumping effect of rhythmical effleurage may be important in oedema removal and that the constant pressure produced by the string wrapping ensured the swelling did not return to the tissues through leaky capillaries. This points to the need to combine oedema massage with some form of pressure to maintain its effects between treatments.

While the results of different studies can appear confusing, it seems that general massage, arbitrarily used, can produce erroneous results. When massage is more specific, however, and carried out sensitively, is anatomically correct and applied at a precise depth, specific strokes can indeed achieve a specific purpose. Clinical experience suggests that a positive effect on physiological parameters is more likely in the presence of pathology. Trubetskoy et al (1997) conducted research which showed that gentle manual massage for 5 minutes increased absorption of subcutaneously injected substances from the tissues into the lymphatics. Effleurage and manual lymphatic drainage can mechanically produce a milking effect or open lymphatic flaps for the removal of proteins. A rhythmical pumping effect can then be achieved. The choice of technique or combination of techniques should therefore be selected carefully, the selection being informed by relevant pathophysiology.

Blood constituent readings following massage can offer further elucidation on the mechanism by which massage works. Arkko et al (1983) conducted research in which vigorous conventional whole body massage, using oil as a lubricant, was carried out for 1 hour in nine healthy male volunteers. Stroking, kneading, friction and shaking were applied by an experienced therapist. Blood samples were taken before, immediately after and at 2, 24 and 48 hours later. A variety of blood and serum constituents were measured and results showed wide individual variation, none reaching statistically significant levels. The results did, however, substantiate the findings of Bork et al (1971) that serum levels of creatine kinase (CK) and lactate dehydrogenase (LDH) were raised. These are enzymes that can be examined for skeletal muscle specificity. Bork and co-workers suggested that LDH was liberated by muscle cells, probably as a result of the mechanical trauma of the massage. This work was not in agreement with the claim by Wood and Becker (1981) that haemoglobin levels and erythrocyte count are raised. Unfortunately, there was no check on the activity levels of the subjects earlier in the day. This group of volunteers was not compared with a control group which did not receive massage. Nine is too small a number to conduct the statistical tests used here with confidence (such as the one-tailed paired t test). A larger group may have shown more (or, indeed, less) significant results and, in comparison with controls, would have increased the validity of this study.

Ernst et al (1987), in a group of normal subjects, found that a standard 20-minute massage treatment reduced the haematocrit, blood and plasma viscosity. The suggestion here is that the fluid immediately surrounding poorly perfused vessels has low viscosity owing to its lack of cells and that the vasodilatation caused by massage nearby creates a need for these almost dormant vessels to be recruited. This may be useful where it is desirable to increase local circulation, for example to promote healing. It can also be a help to athletes whose performance would benefit from increased recruitment of blood vessels. In addition, the mechanical effect of the massage causes a removal of the low-viscosity tissue fluid into the circulation. The study offers further evidence that massage produces a flushing and mechanical effect on the circulation, as the local effects here were detected within the bloodstream.

Research has shown that massage can reduce the incidence of deep vein thrombosis (DVT). Sabri et al (1971) found that the incidence of DVT was reduced by 82% in the massaged limb when compared with the non-massaged limb. Knight and Dawson (1976) demonstrated that the occurrence of DVT in the leg can be reduced by massage of the arm. Both these studies, however, used pneumatic compression devices which simply squeeze the whole limb in a rhythmical manner. This does not mirror a manual massage; rather it mirrors the effect of rhythmical muscle contraction on the muscle pump, which increases venous return. This does not indicate that the reduced incidence of DVT on these occasions was due to massage causing reduced blood viscosity, nor that pneumatic compression devices are superior to regular muscle-pumping exercises in anyone confined to bed or undergoing surgery. A comparative study would be helpful for clarification.