Mekanismene bak utvikling av ødem

Transkript

1 1 Mekanismene bak utvikling av ødem Einar Stranden SE ABSTRAKT BAKERST Sirkulasjonsfysiologisk avdeling Oslo Vaskulære Senter Aker universitetssykehus HF, Oslo Ødem i bena Spreng, ubehag Redusert mobilitet i ledd Redusert lokal nutrisjon Økt sykelighet Vanskeliggjør postop. rehabilitering

2 2 Ødem i bena Tema Transkapillær væskebalanse Kliniske kasus Ødemprevensjon Transkapillær væskebalanse

3 3 Transkapillær væskebalanse Starling: F = CFC((P c -P if ) - σ(cop pl - COP if )) = L Arterie Kapillær Vene Interstitium Transkapillær væskebalanse Starling: F = CFC((P c -P if ) - σ(cop pl - COP if )) = L Veke - i nål - teknikk Arterie P c Kapillær Vene P if P if Interstitium

4 4 Transkapillær væskebalanse Starling: F = CFC((P c -P if ) - σ(cop pl - COP if )) = L Hvordan reguleres kapillærtrykket? Pre- / postkapillært motstandsforhold R a Arterie P c Kapillær Vene R v R a R v Interstitium Transkapillær væskebalanse Starling: F = CFC((P c -P if ) - σ(cop pl - COP if )) = L COP: Osmotisk trekkkraft pga. proteiner (primært albumin) Arterie P c P if Kapillær COP pl COP if Vene Problem: Hvordan skaffe interstitiellvæske? Interstitium

5 5 Transkapillær væskebalanse Fremskaffelse av interstitiellvæske: Blister-metoden Transkapillær væskebalanse Måling av kolloidosmotisk trykk (COP)

6 6 Transkapillær væskebalanse Starling: F = CFC((P c -P if ) - σ(cop pl - COP if )) = L Permeabilitet CFC = K f * SA K f Arterie P c Kapillær COP pl Vene SA P if COP if Interstitium Arterie Strain gauge pletysmografi Bestemmelse av CFC

7 7 Pletysmografisk bestemmelse av CFC mmhg Leggvolum CFC: ml / min. x 100g x mmhg Tid Venøs okklusjon Transkapillær væskebalanse Starling: F = CFC((P c -P if ) - σ(cop pl - COP if )) = L Permeabilitet Ødem etter karkirurgi: +2,0x Stranden 83 DVT: +1,4x Seem og Stranden 90 Arterie P c P if Kapillær COP pl COP if Vene Idiopatisk syklisk ødem: +2,5x Stranden 92 Iskemisk ødem: +2,0x Anvar, Khiabani et al. 99 Interstitium Ødem ved lårhalsbrudd: 1,5x Kazmi,, Stranden et al. 05

8 8 Transkapillær væskebalanse Starling: F = CFC((P c -P if ) - σ(cop pl - COP if )) = L Lymfeåre ca. 10 % Arterie P c Kapillær COP pl Vene P if COP if Interstitium Transkapillær væskebalanse Starling: F = CFC((P c -P if ) - σ(cop pl - COP if )) = L Lymfeåre ca. 10 % Arterie P c Kapillær COP pl Vene P if COP if Interstitium Stranden, Kramer 1982

9 9 Transkapillær væskebalanse Starling: F = CFC((P c -P if ) - σ(cop pl - COP if )) = L Lymfeåre ca. 10 % Arterie P c Kapillær COP pl Vene P if COP if Interstitium Stranden, Kramer 1982 Transkapillær væskebalanse Starling: F = CFC((P c -P if ) - σ(cop pl - COP if )) = L Lymfeåre Trykktransducer ca. 10 % Arterie P c Kapillær COP pl Vene P if COP if Interstitium Stranden, Kramer 1982

10 10 Lymfehjerter Stranden, Kramer 1982 Lymfehjerter Stranden, Kramer 1982

11 11 Ødemmekanismer Lymfeåre Økt filtrasjonskraft Red. kolloidosmotisk gradient Økt permeabilitet Arterie P c Kapillær COP pl Vene Redusert lymfetransport Andre årsaker P if COP if Interstitium Kliniske eksempler

12 12 Økt filtrasjonskraft Red. COP-gradient Økt permeabilitet Redusert lymfetransport Andre årsaker Ødem i bena Økt filtrasjonskraft Red. COP-gradient Økt permeabilitet Redusert lymfetransport Andre årsaker Ødem i bena Dyp venøs s trombose

13 13 Økt filtrasjonskraft Red. COP-gradient Økt permeabilitet Redusert lymfetransport Andre årsaker Ødem i bena Venøs s insuffisiens Økt filtrasjonskraft Red. COP-gradient Økt permeabilitet Redusert lymfetransport Andre årsaker Ødem i bena Høyresidig hjertesvikt Venstresidig hjertesvikt (forplanting til h. side)

14 14 Økt filtrasjonskraft Red. COP-gradient Økt permeabilitet Redusert lymfetransport Andre årsaker Ødem i bena Graviditet 3. trimester Økt filtrasjonskraft Red. COP-gradient Økt permeabilitet Redusert lymfetransport Andre årsaker Ødem i bena Enkelte legemidler Økt permeabilitet

15 15 Økt filtrasjonskraft Red. COP-gradient Økt permeabilitet Redusert lymfetransport Andre årsaker Ødem i bena Lav proteinmengde i blodet (hypoproteinemi) Proteintap Nyresvikt Malabsorbsjon Svangerskapsforgiftning Red. protein-syntese Leversvikt Ernæringsforstyrrelser Malabsorbsjon Økt filtrasjonskraft Red. COP-gradient Økt permeabilitet Redusert lymfetransport Andre årsaker Ødem i bena Allergiske reaksjoner (histaminfrigjøring) ring)

16 16 Økt filtrasjonskraft Red. COP-gradient Økt permeabilitet Redusert lymfetransport Andre årsaker Ødem i bena Brannskade Økt filtrasjonskraft Red. COP-gradient Økt permeabilitet Redusert lymfetransport Andre årsaker Ødem i bena Inflammasjon/infeksjon

17 17 Økt filtrasjonskraft Red. COP-gradient Økt permeabilitet Redusert lymfetransport Andre årsaker Ødem i bena Lymfødem Økt filtrasjonskraft Red. COP-gradient Økt permeabilitet Redusert lymfetransport Andre årsaker Ødem i bena Lipødem Lymfødem

18 18 Økt filtrasjonskraft Red. COP-gradient Økt permeabilitet Redusert lymfetransport Andre årsaker Ødem i bena Ultralyd B-mode skan Lymfødem Lipødem Økt filtrasjonskraft Red. COP-gradient Økt permeabilitet Redusert lymfetransport Andre årsaker Ødem i bena Myksødem Lavt stoffskifte (hypotyreose)

19 19 Hvordan hindre ødem? Aktiver venepumpene! Volumforandring (%/h) Volum Volum p = 0,008 Låst Bevegelig Stolbevegelse Stranden 2000

20 20 Reduser kapillærtrykket! Venøse leggsår Reduser kapillærtrykket! Kurativt +++ Kurativt ++ Kurativt +

21 21 Øk vevstrykket! Elastiske strømper Klasse 1 Lang transportvei P if Elastiske strømper Øk vevstrykket! Elastiske strømper Klasse 1 Lang Kort transportvei P if Elastiske strømper

22 22 Hos eldre: Sørg for adekvat kosthold! Albumin!

23 Hevelse i beina Einar Stranden, Sirkulasjonsfysiologisk avd., Aker universitetssykehus HF, 0514 Oslo einar.stranden@medisin.uio.no I tillegg til dette abstraktet vises til utdelt hefte: E. Stranden, Hevelse i beina Innledning Hevelse i beina er ikke uvanlig. Tilstanden kan være plagsom og i enkelte tilfeller smertefull for pasienten. For legen kan hevelse representere en diagnostisk utfordring, fordi det er en uspesifikk tilstand som man finner ved en rekke sykdommer, fra ubetydelige til potensielt livstruende. Ødem er i seg selv imidlertid ikke ensbetydende med sykdom. For spesielt kvinner er dette en kjent tilstand før menstruasjon og under graviditet. Og mange friske mennesker kan observere hevelse i beina etter langvarig oppreist stilling eller passiv stolsitting. Perifere ødemer har uheldige konsekvenser for pasienten, som for eksempel hindret nutrisjon for cellene (på grunn av øket diffusjonsdistanse), utvikling av hudsår (bl.a. som følge av trangere skotøy), og dårligere bevegelighet i ledd. Kunnskap om de forskjellige mekanismene bak hevelsen muliggjør en rasjonell og systematisk tilnærming til pasientens problem og således en raskere diagnostikk og behandling. Væsketransport i vevet Ødem er opphopning av væske i det interstitielle eller intracellulære volum. I denne presentasjon omtales tilstander som øker interstitiets volum. Interstitielt væskevolum holdes vanligvis relativt konstant, og utgjør ca. 20 % av kroppsvekten. Subkutant ødem oppdages ikke klinisk før interstitielt volum er økt med mer enn 100 %, noe som korresponderer med en økning i totalvolum i størrelsesorden 10 %. I normalt vev er væsken i interstitiet bundet i en «gelé-aktig» konsistens, men blir mer flytende ved økende hydrering. Den transporteres da lettere i vevet, som for eksempel ved dannelse av et søkk etter press mot huden med en finger («pitting» ødem). Uavhengig av bakenforliggende årsak skyldes ødem hovedsaklig en ubalanse mellom filtrasjon og reabsorbsjon i kapillærene, og fjerning av væskeoverskudd via lymfeårene. Væsketransport ut fra blodårene foregår i de svært små og tynnveggete kapillærene. Mengden blod som passerer gjennom kapillærene reguleres ved hjelp av åpning og lukning av små blodårer rikelig forsynt med muskellag i veggen nær kapillærene (Figur 1, m). Denne variasjonen i åpning og lukning påvirker blodtrykket inne i kapillæret kapillærtrykket (Figur 1, Kt). Når blodåren trekker seg sammen reduseres kapillærtrykket, mens det økes når blodåren utvider seg. Dette påvirker også hvor mye væske som filtreres ut fra kapillærene jo høyere kapillærtrykk, desto mer væske vil filtreres ut. Kapillærtrykket er ikke konstant langs kapillæret. I horisontalt leie er det ca. 25 mmhg nær arterien men reduseres til ca. 10 mmhg nær venen. Det aller meste av den væsken som filtreres ut (ca. 90 %) vil imidlertid returnere til kapillæret (reabsorberes) før blodet kommer til venene (Figur 1). Dette skyldes en annen hovedkraft i væskebalansen det osmotiske trykket forårsaket av proteiner. Det finnes mange proteiner i blodet, men det viktigste proteinet for væskebalansen er albumin. Dette osmotiske trykket trekker væske tilbake til kapillærene (Figur 1, Osm). Det finnes riktignok proteiner også utenfor kapillærene i interstitiet. Disse trekker væske i motsatt retning, men konsentrasjonen her er mye lavere enn i blodet (ca. 30 % ved ankelen), og disse har normalt mindre betydning. 1

24 Vi har dermed to hovedkrefter som er sentrale i væskebalansen, kapillærtrykket som bidrar til væsketransport ut av kapillæret, og det osmotiske trykket pga. proteiner som trekker væske tilbake. Normalt filtreres mer væske ut av kapillærene enn det som trekkes tilbake i størrelsesorden 10 %. Når vi vanligvis likevel ikke får ødemer i kroppen skyldes det at dette væskeoverskuddet tas hånd om av lymfeårene. Dette er tynne årer med åpninger i enden der væske kan komme inn. Lymfeårene er rikelig forsynt med klaffer som sikrer transport i kun én retning. I veggen er det muskelceller som rytmisk trekker seg sammen slik at væsken drives fram i lymfeårene («lymfehjerter») 1. Manuell massasje («lymfemassasje») forsterker denne transporten. Figur 1. Skjematisk framstilling av væsketransport i vevet. For nærmere forklaring, se teksten over. Betraktningene over gjelder for en person i horisontalt leie. Væskebalansen kompliseres når vi reiser oss og står i oppreist stilling. Fordi vi befinner oss i et gravitasjonsfelt vil arterier og vener i ankelnivå få et tilleggstrykk på mmhg forårsaket av vekten av blodsøylen fra hjertet og ned til ankelen. Økningen i arterietrykket kan i en viss grad kompenseres ved sammentrekning av arteriolene, men ikke nok til å unngå en massiv økning av filtrasjonen. Når de fleste av oss likevel kan gå rundt uten å få ødemer skyldes dette venepumpesystemene i beina (muskel-venepumpene i leggen, stempelpumpen distalt på leggen og fotvenepumpen) 2. Dette er særdeles effektive mekanismer 3 som ved den minste bevegelse av beina raskt reduserer trykket i venene fra mmhg ned til ca. 30 mmhg. En forutsetning for at disse pumpene skal fungere er intakte veneklaffer i beina (Figur 1, Vk). Vi har flere hundre veneklaffer i leggene, men antallet avtar oppover langs låret, og er nesten fraværende i bekkenet. Dette er et uttrykk for at de viktigste venepumpene ligger nedenfor kneet der venetrykket er høyest i oppreist stilling. Sammenfatningsvis kan nevnes følgende mekanismer som bidrar til dannelse av ødemer: Økt hydrostatisk filtrasjonskraft Redusert kolloidosmotisk trykkgradient Økt kapillær permeabilitet Redusert lymfetransport og følgende ødembegrensende mekanismer: Aktivering av venepumpesystemene Økning av kolloidosmotisk trykkgradient 2

25 Ødemer i beina er ikke en sykdom i seg selv, men ofte et symptom på andre sykdomstilstander. Graden av symptomer varierer fra rent kosmetiske til invalidiserende plager. Beroende på bakenforliggende årsak kan vi dele ødemer inn i følgende hovedgrupper: I Tilstander som øker kapillærtrykket Lokal venøs hypertensjon Vena cava inferior/iliaca kompresjon Dyp venetrombose Kronisk venøs insuffisiens Fascie losje syndrom Systemisk venøs hypertensjon Hjertesvikt Enkelte hjertesykdommer Arterioledilatasjon Enkelte legemidler, for eksempel kalsiumblokkere 4 Øket plasma volum Hjertesvikt Nyresvikt Enkelte legemidler Svangerskap Hormonelle svingninger (premenstruelt ødem) II Tilstander som reduserer kolloidosmotisk trykk i plasma Proteintap Malabsorbsjon Svangerskapsforgiftning Nefrotisk syndrom Redusert protein syntese Cirrhose/leversvikt Ernæringsforstyrrelser Malabsorbsjon III Øket kapillær permeabilitet Allergiske reaksjoner med histaminfrigjøring Brannskade Inflammasjon/lokal infeksjon IV Lymfatisk obstruksjon (lymfødem) Kirurgi, stråleskade Filariasis V Andre årsaker Idiopatisk ødem Myxødem Lipødem Litteratur 1. Stranden E and Kramer K. Lymphatic and transcapillary forces in patients with edema following operation for lower limb atherosclerosis. Lymphology 1982; 15: , Stranden E. Venetømningens fysiologi. Nordisk lærebok «Vensjukdomar». Studentlitteratur, Lund, Sverige Stranden E. Dynamic leg volume in a locked and free floating tilt office chair. Ergonomics 2000; 43(3): Lund-Johansen P, Stranden E, Helberg S, Wessel-Aas T, Risberg K, Rønnevik PK, Istad H and Madsbu S. Quantification of leg oedema in postmenopausal hypertensive patients treated with lercanidipine or amlodipine. Journal of Hypertension 2003: 21:

26 I artikkelen under beskrives hvor utrolig effektivt muskel-venepumpe-systemet i bena fungerer. Dette forutsetter at veneklaffene ikke er skadet og at man ikke har venøs insuffisiens.

27 421 ERGONOMICS, 2000, VOL 43, NO. 3, Dynamic leg volume changes when sitting in a locked and freefloating tilt office chair EINAR STRANDEN* Department of Vascular Diagnosis and Research, Surgical Clinic, Aker University Hospital, 0514 Oslo, Norway Keywords: Oedema; Oedema prevention; Office chairs; Dynamic sitting; Plethysmography It is well established that prolonged sitting may lead to swelling of the lower extremities. However, activation of the vein pump system by repeated walking breaks or dynamic tiltable foot-rests have been shown to reduce foot and leg oedema. Some office chairs incorporate tilt-mechanisms facilitating movements of the body from the feet up. The present study was undertaken to establish whether a beneficial effect on the transcapillary fluid balance of the legs by enabling such mechanisms could be documented. An office chair where the tilt mechanism could be locked or open (HÅG Credo 3500) was used for the study. The seat position and seat activity level was recorded by a transducer-system developed for the study. Calf volume and calf muscle pump activity was detected by mercury strain gauge plethysmography. In the locked position there was a steady increase in volume of mean 1.2 % (range ) for all subjects in the 30 min. study period. On the other hand, for all subjects there was a decrease in calf volume (mean 0.7 %, range , p=0.008) when the tilt-mechanism was open (30 min. period), irrespective of what study period came first. The study showed that upward seat deflection was not associated with concomitant venous obstruction, since such obstruction was detected in less than 2 % of the time period with more than 50 % of maximal upward deflection. A locked seat mechanism does not prevent activation of the vein pump mechanism, but the study indicates that office chairs that permit variation in seat angle per se stimulate movements of the leg. This, in turn, activates vein pumps and counteracts local oedema formation in seated working postures. 1. Introduction An increasing number of jobs are associated with sitting postures (Peters 76, Östberg 77, Winkel 81). However, prolonged sitting may lead to swelling of the lower extremities, especially in constrained postures (Gauer and Thron 1965, Pottier et al. 1969, Winkel 1981, Stranden et al. 1983, Paul 1995, Stranden and Kroese 1998). Oedema results from increased net transcapillary filtration, exceeding removal of fluids by the lymphatics. The dominant factor promoting oedema is increased capillary hydrostatic pressure, caused by increased venous pressure. The vein pressure of the feet in a horizontal position is approximately *Correspondence: einar.stranden@ioks.uio.no

28 422 5 mmhg, increasing to mmhg in a upright position. At passive sitting, the pressure is in the order of mmhg, determined by the height of the blood column from the feet to the heart. When walking, the vein pressure normally decreases to mmhg, caused by vein pump mechanisms in the lower extremities (Stranden et al. 1986, Gardner and Fox 1989). Based on this knowledge, Winkel (1981) and Winkel and Jørgensen (1986) suggested inclusion of short walking breaks in office work to reduce the amount of leg/foot oedema. Significant reduction in swelling of the legs has also been documented, when using a specially designed dynamic foot-rest in constrained factory workstations. This was based on a pivot that allowed "treadle" action to be performed by the operator (Stranden et al. 1983, Stranden and Kroese 1998). Some office chairs incorporate a tilting-mechanism that combines the movement of the seat and backrest in such a way that the front of the seat is lifted upwards when one leans on the backrest. Among the rationales for such a mechanism is that they facilitate movements of the whole body from the feet up (dynamic sitting pattern). Consequently, one may assume that if the muscle activity of the lower extremities is stimulated, this would enhance venous pumping, thereby reducing the distal vein hydrostatic pressure. Critics of such a tilt mechanism argue that a seat deflection upwards when leaning on the backrest results in increased pressure at the back of thigh and thus causing venous obstruction and increased transcapillary filtration. The current study was undertaken to investigate whether the dynamic sitting pattern of office chairs with free-floating tilt mechanism had significant beneficial effect on transcapillary fluid balance in healthy controls Physiological considerations Lower extremity vein pump system: The most important oedema limiting mechanism of the lower extremities is the venous pumping system, which, when activated, reduces distal venous pressure at upright or seated position. In relation to function, the venous pumping system may be divided into three portions with different working mechanisms (Gardner and Fox 1989) (figure 1): (1) the muscle pumps; (2) the distal calf ("piston") pump; and (3) the foot pump The muscle pumps: The pump unit consists of muscles ensheathed by a common fascia, which is drained by a set of densely-valved intra- and intermuscular veins. These in turn empty into more sparsely valved proximal veins (Arnoldi 1989). The leg contains four muscular compartments: The anterior, lateral, deep posterior and superficial posterior, all drained by their respective veins. Muscle contraction is the main activator of muscle pumps (Ashton 1975), but passive stretching may also raise intramuscular pressure and promote pumping. Baumann et al. (1979) recorded pressures of >100 mmhg in the tibial anterior muscle during contraction and 35 mmhg on passive stretching. Contraction of the calf muscles initiates a rise in pressure in all veins of the lower limb. The increase is most pronounced within muscle veins, three times higher than in superficial veins. During muscle contraction (systole) the greatly increased pressure difference between deep calf veins and the popliteal vein causes rapid flush of blood from the calf to the thigh (figure 1). Retrograde flow is prevented by competent venous valves. On subsequent muscle

29 Figure 1. Schematic illustration of the venous pump systems of the foot and calf in relaxed and active state. The muscle pump unit consists of muscles (M) ensheathed by a common fascia (F) and veins within the same compartment. Contraction of the calf muscles (muscle systole), as in plantar flexion of the ankle joint during walking (below), expels blood into the proximal collecting vein. During relaxation (muscle diastole, above) the blood is drained from the superficial veins (SV) into the deep veins (DV) in addition to the arterial inflow, making the pump ready for the subsequent ejection. V: venous valve. The distal calf pump is indicated in the middle. On dorsiflexion of the ankle (passive or active), the bulk of the calf muscle (M) descends within the fascial sheath (F), and expells blood in the distal veins like a piston. The foot vein pump is illustrated to the right. The plantar veins are connected like a bow-string from the base of the fourth metatarsal in front to the medial malleolus. On weightbearing the tarso-metatarsal joints are extended and the tarsal arch is flattened. Thus the veins are stretched, causing them to eject their content of blood. 423

30 424 relaxation (diastole) venous pressure falls below the pressure at rest. The fall is greatest in the deep veins, less in the superficial veins and insignificant in the popliteal vein. In this phase perforator veins allow flow from the superficial to the deep veins, whereas competent valves prevent backflow from the popliteal to the deep calf veins. The calf pump is probably the most important muscle pump. However, also the thigh pumps (quadriceps muscle pump, sartorial muscle pump, the pump of the hamstring muscles) and the popliteal vein pump (Hach 1976) play a part in the centrally directed propulsion of blood The distal calf ("piston") pump: In contrast to conventional descriptions, there are two pumping systems in the calf, a proximal and a distal (Gardner and Fox 1989). The distal one is activated on dorsiflexion of the ankle (figure 1), when the calf muscles are stretched and their distal part descends within the fascial sheath. This movement acts like a piston which expels venous blood in proximal direction. The pump mechanism has been documented by ultrasound Doppler measurement of venous blood flow (Gardner and Fox 1989), and is supported by compartment pressure measurements (Gershuni et al. 1984) The foot pump: The significance of a pumping system within the foot has often been overlooked, although first postulated by Le Dentu in Gardner and Fox (1983) used video-phlebographic technique and ultrasound Doppler measurement to demonstrate a potent pump mechanism in the deep plantar veins. The pumping mechanism has been explained as follows. The plantar veins are connected like a bow-string between the base of the fourth metatarsal and the medial malleolus. On weight-bearing, the tarso-metatarsal joints are extended and the tarsal arch is flattened. Thus the veins are stretched, causing them to eject their content of blood (figure 1). The pump is also activated on weight-bearing of the forefoot alone, when the foot is acting like a lever (Gardner and Fox 1989) (figure 1, left portion) The pumps acting together: During normal walking the three vein-pumping systems are synchronised to form a complete network of serial and parallel pumps aiding the return of blood towards the heart. Even moderate muscular movements of the legs in seated position may activate the pumping mechanism, significantly reducing distal mean vein pressure (Stranden 1981, Stranden et al. 1983, Stranden and Kroese 1998) 2. Methods 2.1. Participants Eight healthy control subjects (3 females, 5 males) were studied. The mean age was 35 years (range 21-48). None had clinical signs or medical history indicating arterial or venous disease Chair An office chair with a centre tilt mechanism and a fixed angle between the seat and the back was used for the study (HÅG Credo 3500, HÅG a.s.a. Oslo, Norway). The linkage between the seat and the back results in an upwards deflection of the seat front when the chair is tilted backwards and the seat front is moved downwards when the chair is tilted forwards. The chair had adjustable seat height and seat depth as well as adjustable supporting point of the backrest. The backrest also had an individual pivot action. Resilient means are attached

31 425 on either side of the pivot point of the seat. These could be adjusted to fit the users preference and ensure balanced sitting. This chair was chosen because of its wide spread use and because the tilt mechanism was easily locked/unlocked with a lever. When the tilt mechanism is open the movement that occurs is described as "free-floating tilt" 2.3. Seat angle/activity detection For continuous recording of seat angle and seat activity, a transducer system was developed (figure 2). The transducer detects the position of the seat/backrest relative to the fixed, vertical axis of the chair (90 to the floor). The detector element is a potentiometer supplied with a ± voltage. The detector output voltage is balanced at zero when the chair is neutral, positive when the seat front is deflecting upwards, negative when deflecting downwards. The signal was fed via an analogous/digital converter (DT2801, Data Translation, Inc., Marlboro, MA, USA) to a PC. Figure 2. The transducer system developed (E.S.) for the detection of seat angle/activity. The iron bar A is fixed to the vertical chair stem, and B follows the movements of the seat/backrest. The wire/spring (s) attachment to a potentiometer (P) ensures a variable output voltage, representing seat deflection. The potentiometer is supplied with a ± voltage (PS). The detector output voltage is balanced at zero when the chair is neutral, positive when the seat front is deflecting upwards, negative when deflecting downwards.

32 Measurement of leg volume changes Variation in calf volume was measured by strain-gauge plethysmography (Whitney 1953), using a double mercury in silastic strain-gauge around the greatest circumference of the calf. Indentation of the skin underneath the strain-gauges was prevented by underlying matches. The plethysmograph ( Strain Gauge Plethysmograph, developed by E.S.) was equipped with electrical calibration (Brakkee and Vendrik 1966) permitting repeated calibrations between measurements. The electrical calibration was compared with mechanical stretching of the strain-gauges by use of a micrometer, and an exact linear relationship was found between extension of the gauges and the recorded volume change Recording of data Perisoft software (Perimed AB, Järfälla, Sweden, was used for recording and analysis of data. The recorded data were: 1. Variation in calf volume, and 2. Seat angle/activity. The software allowed changes in timescale (x-axis) and signal amplification (y-axis) to fascilitate the analyses. The software included Fourier analysis as described by Press et al. (1997) Experimental design Following connection to the measurement devices, the chair was adjusted individually, and it was also ensured that the thigh was in a horizontal position with 90 flexion of the knee joint. A fixed angle of the knee in the resting condition before and after measurement was important, because the cross section of the passive calf muscles is dependent upon the muscle extension. The calf volume level before and after each session was determined following a 2-min.-period with the legs at rest, to ensure blood-filled veins. Calibration signals (1.0 % calf volume increase) were included at the beginning and end of each recording. The recordings were performed under the following two conditions: 1. Tilt-mechanism in locked position for 30 minutes (locked phase). The person was free to move, but the chair did not follow the movement. 2. Tilt-mechanism open, allowing the free-floating tilt of the chair for 30 minutes (freefloating phase). In half of the studies the procedure was carried out in reversed order, to establish whether the procedure itself affected the measurements. The study leg was chosen at random. During the two phases, the study persons were occupied with standard seated office tasks, manipulating documents on the desk and performing computer procedures. Before measurements, they were familiarized with the chair in a test period. They were instructed to choose their own preferred sitting pattern and leg activity level 2.7. Statistics With the limited number of persons included, normal distribution was not expected, and nonparametric tests were chosen. For comparison between the investigation phases, Wilcoxon signed rank sum test (two-tailed) for paired observations were used (program Graphpad Instat 3.0, Graphpad Software, Inc., San Diego, CA, USA, Differences were considered significant when p-values were less than 0.05.

33 Results A typical recording of leg volume change and seat activity during the two study phases is shown in figure 3. The locked phase is shown to the left. Relatively little muscle movements are recorded, as indicated by the steadily increasing volume curve with minor deflections. To the right the tilt-mechanism is open, allowing free-floating tilt. A clearly more active sitting pattern follows, indicated by the volume curve and the transducer signal. Secondly, a slight downward volume slope by time may be drawn in this portion of the measurement. In figure 4 the volume changes for all extremities in the two phases are depicted, expressed as volume change per study period of 30 min. For all there was a volume increase in the locked phase (mean 1.2 %/30 min., range ), and varying degree of volume reduction in the free-floating phase (mean 0.7 %/30 min., range ). The difference between the phases was statistically significant (p=0.008). In two subjects the volume recording during free-floating was two-phasic, consisting of both volume decrease and increase. For these, the net change in volume between start and end of the 30 min. period was used, making these results less reliable. No obvious difference in movement pattern between these two and the others was detected. Figure 3. Typical recording of calf volume changes (upper curve) and chair seat activity as detected with the transducer described in figure 2 (lower curve). The left portion of the recording represents the study phase with the centre tilt mechanism locked; to the right the seat is made tiltable by enabling the centre tilt mechanism. Each period has a duration of 30 minutes. The transient changes in volume, indicated by A, represent changes in the cross-section of the muscles underneath the strain gauges taking place when varying the degree of flexion of the knee joint. The changes do not represent variation in venous- or interstitial volumes.

34 428 Change in calf volume (% / 30 min.) p = Locked Free floating Figure 4. Calculated volume changes (%) per 30 min. with the seat locked (volume increase) and open (volume reduction) for the eight measurements. There was a significant difference between the two phases. There was no significant difference between the recordings that started with the seat locked and those that started with the free float phase. Figure 5 represents a 5-min. selection from the free-floating phase in figure 3. The volume curve indicates volume reductions by activating the vein pumps at even moderate seat deflections (left portion of the curve). Following each deflection, the volume increased steadily during the next 30 s to approach the pre-deflection level, as the veins are filled from the arterial side. Increasing the level of seat deflection (right portion of the curve) does not imply corresponding increase in calf volume deflections. This finding is even more prominent in figure 6, showing volume changes of the same order of magnitude as those in figure 5, at minimal seat deflections. Similar recordings were also found for the other subjects. As the study did not show unambiguous correlation between vein function and seat activity, the recordings were reanalysed, focusing on venous pumping, irrespective of seat movements. In this analysis significant venous pumping was recorded, defined as leg activity causing volume reduction of more than 0.5 %. The result from this analysis is shown in figure 7. There was an increase in the number of significant vein pump events when the chair had a free-floating tilt, from 50 events per 30 min. in the locked phase (range 23-82) to approximately 200 per 30 min. in the tiltable phase (mean 191, range ) (p=0.008). To establish a possible relationship between upward seat deflection and venous obstruction, the periods with more than 50 % upward deflection was examined. All subjects together, the sum of these periods amounted to about 85 min. (range 3-17 min.) of a total free-floating study time of 240 min. In these periods, only four incidences for a total of 1.5 min., with a slight volume increase indicating venous obstruction was detected.

35 429 Figure 5. A 5-min. selection from the free-floating tilt phase in figure 3, indicating moderate seat deflection in the left portion and more extensive seat movements to the right. Note that the venous pumping, as judged from the downward slopes (ejection of venous blood) not necessarily becomes larger when the magnitude of the seat movements are intensified. Figure 6. A 5-min. selection from the free-floating tilt phase in figure 3, showing venous ejections with similar magnitude as in figure 5, at minimal seat deflections. The ejections are predominantly based on static muscular activity. The great variation in movement pattern for the subjects is indicated in figure 8, a Fourier analysis of the free-floating tilt- and locked seat phases. Major movement components are in the frequency domain below 10 cycles/min., but with large interindividual differences. The frequency plots from the tilt phases were subjectively graded in a scale of 1-5, according to amplitude and frequency distribution. There was no correlation between these data and the

36 Calf pump activity (Number of >0.5 % ejections) p = Locked Free floating Figure 7. Number of significant venous pump manoeuvres in the locked and free-floating tilt phases, defined as leg activity causing volume reductions of more than 0.5 %. The increase in muscular activity level was statistically significant. level of calf volume reduction. In fact, the two recordings with the highest score were among the three with the lowest volume reduction in the active phase (mean 0.3 %). Furthermore, for the two subjects with the lowest score the volume reduction was above the mean value for all subjects (mean 0.7 %). 4. Discussion The most important finding in this study was that use of a seat attachment mechanism allowing free-floating tilt movements stimulates sitting patterns that activate vein pumps of the leg. Previous studies (Stranden et al. 1983, Stranden and Kroese 1998) showed that volume reduction during activation of the vein pumps occur in parallel with reduction in vein pressure. In periods with a large number of significant vein pump manoeuvres the mean venous pressure is presumably reduced. This reduces capillary pressure, which is the most important component for altering transcapillary fluid balance, thereby reducing net transcapillary filtration. Another finding was that functional vein pump manoeuvres did not presuppose variation in seat angle, even in the free-floating tilt phase. Hence, at least part of the physiological venous pumping is caused by static muscle activity. This type of muscular movements is present in both study phases. There is no reason to believe that locked seat position per se prevents these static contractions. As the analyses nevertheless showed three times higher frequency of significant pumping manoeuvres in the tilt phase, an interpretation may be that this seat mechanism and combination of fixed angle between seat and backrest stimulates movements, since the working tasks in the two phases were identical.

37 431 Figure 8. Fourier frequency analysis of the two study phases for each leg studied. The diagrams for the locked phase are truncated because there was no detected activity above the frequency domain 0-20 cycles/min. A great variation in movement pattern between the individuals is evident from the free-floating tilt phase. The present seat mechanism that incorporates movements like those of a rocking chair is suitable for activating the vein muscle pumps. The plantar flexion of the feet and contraction of the calf muscles taking place when leaning back effectively initiates venous pumping (Stranden et al. 1983). The degree of calf volume reduction did not depend on the extent of leg movements, as evaluated by the frequency analysis. The interpretation may be that the venous pumping mechanisms are very potent and effective, and venous drainage causing reduction in venous pressure occurs even at moderate, but frequent movements. This is in line with a previous study where we found that a single plantar flexion was as effective in

38 reducing venous pressure as a series of pumping manoeuvres (Stranden et al. 1983). Another explanation lies in interindividual differences between the subjects studied. We have previously shown that constrained working postures often lead to passive sitting and oedema formation in foot and leg (Stranden et al. 1983). The fluid accumulation in the tissue resulted in significant increase in tissue hydrostatic pressure, which may partly explain the discomfort reported (paresthesia, numbness, pain). By introducing specially built dynamic footrests allowing treadle action of the feet against a weak counterpressure, the discomfort and swellings were significantly reduced (Stranden et al. 1983, Stranden and Kroese 1998). The present study confirms the previous findings, demonstrating the potential of the venous pumps in restoring normal transcapillary fluid balance in a sitting posture. The volume changes in the current study (both increase and reduction) are somewhat larger than those associated with the dynamic foot-rest. This may be attributed both difference in study population, and the possibility that the freely movable sitting posture when using free-floating chair seat is more effective regarding vein pump mechanisms. Furthermore, the present work tasks were more varied, enabling more freely movement of the body than in the constrained industry workplace (Stranden et al. 1983). The study did not support the finding of Pottier et al. (1969) that upward seat deflection is associated with venous obstruction in the distal thigh. In less than 2 % of the time period with significant (>50 %) upward slope there was a slight volume increase, indicating venous obstruction. A considerable pressure at distal thigh, where the deep vein runs deep within the Hunter s canal, is required to compress the vein, probably far beyond discomfort. Furthermore, the great saphenous vein (superficial vein), which has a great capacity for venous drainage, is located antero-medially and is not subject to compression in seated posture. The only region where compression is likely to occur is in the popliteal area. As the seat depth of study chair is adjustable, the depth of the seat was checked, and when neccessay adjusted, to avoid compression at the knee level in those subjects with a short thigh length. Furthermore, the flexion of the knee joint during upward seat deflection does not represent a likely cause for venous compression. This is evident from earlier studies on Balans chairs, where venous pressure recordings at very much larger flexion did not indicate any venous obstruction at all (Stranden 1981). The most extreme sitting posture was applied at Balans Skulptur, where flexion was about ; still with no venous obstruction affecting venous pressure profiles. The volume recordings during the present 30 min. periods may not be extrapolated to cover a total working day. Usually the working period is interrupted by breaks and walking to different tasks, which has been shown to reduce the oedema development (Winkel and Jørgensen 1986). In conclusion, the present study indicates that office chairs which permit variation in seat angle when the upper body moves per se stimulate movements of the leg. Leg movements are beneficial because they activate the venous pumping mechanisms, thus counteracting local oedema formation in leg and feet, and reduce discomfort in seated working postures. 432

39 433 References ARNOLDI C.C. Physiology and pathophysiology of the venous pump of the calf. In: Eklöf B., Gjöres J.E., Thulesius O., Bergqvist D. Controversies in the management of venous disorders. London: Butterworths, 1989: ASHTON, H. The effect of increased tissue pressure on blood flow. Clin Orthop 1975; 113: BAUMANN, J.U., SUTHERLAND, D.H. AND HANGGI, A. Intramuscular pressure during walking. Clin Orthop 1979; 145: BRAKKEE, A.J.M. AND VENDRIK A.J.H. Strain gauge plethysmography: Theoretical notes on a new design. J Appl Physiol 1966; 21: GARDNER, A.M.N. AND FOX, R.H. The venous pump of the human foot. Bristol Med-Chir J 1983; 98: GARDNER, A.M.N. AND FOX, R.H. The return of blood to the heart: venous pumps in health and disease. London: John Libbey & Co. Ltd GAUER, O.H. AND THRON, H.L. Postural changes in the circulation. In W.F. Hamilton (Ed.) Handbook of Physiology, Vol. III (pp ). Washington, D.C.: American Physiological Society, GERSHUNI, D.H., YARU, N.C., HARGENS, A.R., LIEBER R.L., O HARA, R.C. AND AKESON, W.H. Ankle and knee position as a factor modifying intracompartmental pressure in the human leg. J Bone Joint Surg 1984; 66A: HACH, W. Der Phlebographie der Bein und Beckenvenen. Constanz: Schnetzor-verlag GMBH, PAUL, R.D. Foot swelling in VDT operators with sitting and sit-stand workstations. In: Proceedings of the Human Factors and Ergonomics Society 39th Annual Meeting, Santa Monica, CA: Human Factors and Ergonomics Society PETERS, T. Arbeitswissenshaft für die Büropraxis. Friedrich Kiehl Verlag, Ludwigshafen, POTTIER, M., DUBREUIL, A. AND MONOD, H. The effects of sitting posture on the volume of the foot. Ergonomics 1969; 12: PRESS, W.H., TEUKOLSKY, S.A., VETTERLING, W.T. AND FLANNERY, B.P. Numerical Recipes in C. The art of scientific computing. Cambridge University Press, Cambridge, STRANDEN E. The influence of "Balans" chair sitting on ankle vein pressure. International Design Exhibition New York Abstract. STRANDEN, E., AARÅS, A., ANDERSON, D.M., MYHRE, H.O. AND MARTINSEN, K. The effect of working posture on musculoskeletal load and circulatory condition. In: Coombs, K. (ed.) Proceedings of the Ergonomics Society's Conference Taylor & Francis Ltd., London , STRANDEN E., ØGREID, P. AND SEEM, E. Venous pressure gradients in controls and in patients with chronic venous disease. Phlebology 1, 47-50, STRANDEN, E., KROESE, A.J. Venodynamics in healthy subjects and in patients with venous dysfunction. Scope on phlebology and lymphology 1998; 5: WHITNEY RJ. The measurements of volume changes in human limbs. J Physiol 1953; 121: WINKEL, J. Swelling of the lower leg in sedentary work a pilot study. Journal of Human Ergology 1981; 10: WINKEL, J. AND JØRGENSEN, K. Evaluation of foot swelling and lower-limb temperatures in relation to leg activity during long-term seated office work. Ergonomics 1986; 29: ÖSTBERG, O. Från arbete till operatör. Technical Report, SOT, Department of Human Work Sciences, University of Sweden (in Swedish) 1977.