CA1308982C - Device for applying compressive pressures against a patient's limb - Google Patents

Abstract

A DEVICE FOR APPLYING
COMPRESSIVE PRESSURES
AGAINST A PATIENT'S LIMB
ABSTRACT OF THE
DISCLOSURE
A device for applying compressive pressures against a patient's limb comprising a stocking having a circumferentially elastic boot portion which applies a compressive pressure against the limb which decreases from the ankle to a top of the stocking, an elongated pressure sleeve for enclosing a length of the patient 18 limb over the stocking, said sleeve having a plurality of separate fluid pressure chambers progressively arranged longitudinally along the sleeve from a lower portion of the limb to an upper portion of the limb proximal the patient's heart relative to said lower portion,.
a device for intermittently forming a plurality of fluid pressure pulses and a device for connecting the pressure pulses to chambers in the sleeve to apply a compressive pressure against the patient's limb by the sleeve which decreases from the lower to upper portions.

Description

1 3~

BACKGROUND OF THE INVENTION

The present invention relates to therapeutic and prophylactic devices, and more particularly to devices for applying compressive pressures against a patient' 8 limb.
It is known that the velocity of blood flow in a patient's extremities, particularly the legs markedly de-crease~ during confinement of the patient. Such pooling or stasis of blood is particularly pronounced during surgery, immediately after surgery, and when the patient has been con-fined to bed for extended periods of time. It is also known that stasis of blood is a significant cause leading to the formation of thrombi in the patient's extremities, which may have a severe deleterious effect on the patient, including death. Additionally, in certain patients it is desirable to move fluid out of interstitial spaces in extremity tissues, in order to reduce swelling associated with edema $n the extremities.
The problem of postoperative deep venous thrombosis ~DVT) and prophylactic regimens in its management have been described by the ~ational Institutes of Health Concensus Development Conference on Prevention of Venous ~hrombosis and Pulmonary Embolism. Clearly it is a problem of ma~or concern, and several prophylactic modalities are available to help pre-vent its occurrence.
Modalities which have been identified to be effective in the reduction of postoperative DVT have been categorized on the basis of their mechanism of action in either preventing the hypercoagulable state or preventing sta~is. While anticoagulants have been shown to be ef~ective, they carry a ris~ of bleeding and wound hematoma. On the other hand, complications have not been associated with use of compression modalities, such as intermittent pneumatic compression (IPC) disclosed in U.S.

pabent 4,013,069 and graduated oompression stockings disclosed Ln U.S. patent 3,728,875.
Comblnatlons of prophylactic modalitie~ to act on more than one component of Virchow'~ Triad have been utilized to achieve lncreased prophyla¢tic effectiveness. The additlon of graduated elastic compress10n to low dose heparin reduces the incidence of DVT compared to low dose heparin alone. The addition of dihydroergotamine to low dose haparin has been demonstrated to be more effective in reducing DVT than low dose heparin alone. The finding that certain combinatlon prophylactic regimens are more effective than single modality regimens was indicated in a recently published meta-analysis of the llterature which also emphasized the finding that the combination of graduated compression stockings before or after use of intermittent pneumatic compression (IPC) was more effective than IPC alone.
The foundation of this conclusion involves the original series of Nicolaides on IPC prophylaxis which indicated that IPC
was as effective a8 low dose heparin for the time it was applied;
however, after IPC was discontinued prophylactic protection diminished. Nicolaides AN, Fernandes JF, Pollock AV. Intermlttent ~equential pneumatlc compre~sion of the leg~ in the prevention of venous stasis and postoperative deep venous thrombos~s. Surqery~
87:69-76. 1980. Further work conducted by Nicolaides subsequently combined graduated elastic compression after IPC to provide a "continuity" of prophylaxis 80 that when IPC wa~ discontinued, graduated elastic compression stockings were applied and worn through the remainder of hospital stay. This combination regimen utilizing graduated elastic compression before and after the application of IPC (but not during use of IPC) indicated a reæult comparable to that of low dose heparln. One point demonstrated in this work on IPC i8 that effective prophylaxis requires a -"continuity" of prophylaxis for the entire time the patient is at risk.
The physical methods of prophylaxis, including graduat~d elastic compression and IPC, have long been considered to act by promoting venous blood flow and thereby reducing the stasis component of Virchow' 8 ~riad. The actlon of IPC ha~
been demon~trated to significantly lncrease blood flow pul~atility and enhance blood clearance from the soleal sinuses, the axial veins and the valve sinuses. More recently, it has been indicated that IPC stimulates fibrinolytic activity and, in addition, enhances prostacyclin generation.
Therefore, the prophylactic effectiveness o IPC is thought to arise from potentially two actions, a reduction of venou~ stasis by increasing venous flow pulsatility and reducing the hypercoagu~able state. ~he prophylactic effectlveness of graduated elastic compression stockings is thought to be due primarily to its reduction of venous stasis by increasing linear blood flow velocity.
Previous efficacy studies on IPC as applied to the surgical patient have not used graduated elastic compression stoc~ings simultaneously with IPC, even though some studie~
have u~ed stoc~lngs se~uentlally wlth IPC to provlde an lmproved contlnuity o prophylaxis.

.
.

_a_ z SUMl~LARY OF THE PRESENT Il~VF.l~TIO~
A principal feature of the present invention is the provision of an improved device for applying compressive pressures against a patient's limb.
According to one aspect of the invention there is provided, a device for applying compressive pressures against a patient's limb, comprising in combination a stocking S having a circumferentially elastic boot portion which applies a compressive pressure against the limb which decreases from the ankle to a top of the stocking, an elongated pressure sleeve for enclosing a length of the patient's limb over the stocking, said sleeve having a plurality of separate fluid pressured chambers progressively arranged longitudinally along the sleeve from a lower portion of the limb to an upper portion of the limb proximal the patient's heart relative to said lower portion means for forming a plurality of fluid pressure pulses, and means for COMeCting the pressure pulses to chambers in the sleeve to apply a compressive pressure against the patient's limb by the sleeve which decreases from the lower to upper limb portions.
A feature of the present invention is that the device is a more effective prophylactic regimen than stockings or intermittent pneumatic compression as mentioned above.
Further features will become more fully apparent in the following description ofthe embodiments of this invention and from the appended claims.
DFSCRTl~ION OF T~IE DRAWINGS
In the drawings:
Fig. 1 is a perspective view of a pair of compression sleeves used in the sequential intermittent compression device of the present invention;
Fig. 2 is a front plan view of a compression sleeve of Fig. l;

4a 13~

Fig. 3 is a back plan view of the sleeve of Fig. 2;
Fig. 4 is a sectional view taken substantially as indicated along the line 4-4 of Fig. 3;
Fig. 5 is a schematic view of a manifold for use in connection with the device of Fig. l;
Fig. 6 is a perspective view of the manifold for use with the device of Fig. l;
Fig. 7 is a sectional view taken substantially as indicated along the line 7-7 of Fig. 6;
Fig. 8 is a graph illustrating pressure-time curves during operation of the compression device~
Fig. 9 is a schematic diagram of one embodiment of a pneumatic control circuit for the compression device;
Fig. 10 is a schematic diagram of another embodiment of a pneumatic control circuit for the compression device;
Fig. 11 is a schematic diagram of another embodiment of a pneumatic control circuit for the compression device~
Fig. 12 ahd 13 illustrate ~espeotively a typical stocking view from the inner leg side and the front:
Fig. 14 illustrate the top portion o~ a preferred form , ofcircular knitstocking of the invention~
Fig. lS and 16 illustrate views of the top of a typical stocking before and after the upper thigh circumference has been adjusted;
Figs. 17-20 illustrate typical elastic fabric~ including elastomeric yarns suitable for the boot portion of the stocking:
Fig. 21 i8 a perspective view of a pair of compre~sion sleeves used in the sequential intermittant compression device and a pair of stockings beneath the sleeves of the present invention; and Fig. 22 is a histogram showing the day on which DVT was first detected in a stocking leg and a non-stocking leg.

-4b-~3(~ Z

DESCRIPTION OF THE PREFERRED EMBODIMEN~S
Referring now to Figs. 1, 6, and 9-11, there ~s shown a sequential intermittent compression device generally designated 20 for applying compress,ive pre~sures agAinst a patient's extremities, such as the legs. ~he devioe 20 has a controller 22, as illustrated in Figs. 9-11, a manifold 24, as shown ln Fig. 6, and a pair of compression sleevos 26 for enclosing lengths of the patient's leg~, as shown in Fig.
1. ~he controllers 22 of F$gs. 9-11 interm$ttently form a plurality of fluid prefisure pulses rom a source S of pres~ur-ized ga~ in a timed sequence during periodic compression or inflation cycles, and the pul~e~ are separately applied to ~3~ 2 the manifold 2~ of Fi~. 6 through conduits 28a, 28b, and 28c at inlet port~ of the manifold 24. The man$fold 24 of Fig. 6 separates the pulses for passage to the ~eparate sleeves 26 through two sets of condults 34a snd 34b wh$ch are ~eparately connected to the sleeves, as shown in Fig. 1.
As shown in Figs. 2-4, the sleeves 26 have a pair o~ flexible sheets 36 and 38 which aro made Srom a flu$d $m-pervious material, such as polyvinyl chloride. The sheet~ 36 and 38 have a pair of side edges 40a and 40b, and a pair of end edges 42a and 42b connecting the side edges 40a and b. As shown in Figs. 3 and 4, the sheets have a plurality of laterally ex-tending lines 44, such a~ lines of sealing, connecting the sh-ets 36 and 38 together, and a pair of longitudinally extending lines 46, such as line~ of sealing, connect~ng tho sheets 36 and 38 togother and connecting ends of tho lateral llnes 4i, as shown.
~he connecting line~ 44 and 46 define a plurality of cont$guous chambers 48a, 48b, 48c, 48d, 48e, and 48f which extend lateral-ly ~n the sheet, and which are disposed longitudinally in the sleeve between the end edges 42a and 42b. When the sleeve i8 placed on the patient's leg, the lowermost chamber 48a i~ lo-CA~d on a low-r part of the leg ad~scent tho patient's anklo, wh$1e the uppermo~t chamber i8 located on an upper part of tho leg ad~acent the m$d-thigh.
In a preferred embodiment, the side edges 40a and 40b and the connecting lines 46 are tapered from the end edge 42a toward the end edge 42b. Thus, the sleeve 26 hn~ a rë-duced configuration ad~acent its lo~r end to facilitate place-13f '~?2 mont of the sleeve oh the more narrow reglon~ of the legad~acent the patient?s anklos Moreover, lt wlll bo soon that the connecting lino~ 44 and 46 d-fine chambor~ havlng volumes whioh progressively ~ncrea-e in ~ize from the lower-most chamber 48a to the uppermost ch~mber 48S Th- relativo aize of the chambers fac$1itate~ the device in con~unction with orifices to develop a compressive pressure gradient during the compress~on or inflation cycle~ which decreases from a lower part of the sleeve ad~acent the end edge 42b toward an upper ~art of tho leeve ad~acent tho ond edge 42a Aa lllustrated in Figs 3 and 4, the ad~oining cham-bera 48c and 48d may have their ad~acent portion~ defin-d by spaced connecting lines 44' and 44" which extend laterally in the sleeve between the connecting linea 46 The sheets 36 and 38 may be severed, ~uch a~ by slitting, a~ong a line 50 between the lines 44' and 44" to separate the ad~oining chambers 48c and 48d As shown, the severence lin- 50 may extend the width oS th- chamber~ betwe-n the connectlng lino- 46 Tho line 50 permits free relative mov-ment between the ad~oining chamber~
when the sleeve is inflated to prevent hyperextension of the 13~ Z`

leg during opera~io~ of the device, and also facilitates siz-ing of the 81eeve ti the leg of a particular patient.
The sleeve 26 may have one or more sheets 52 o~ a soft flexible materlal for covering the outsldo of the flu~d impervious sheets 36 and 38 relative the patient's leg. The sheets 52 may be made.of any suitable material, such as Tvvek, a trademark of E.I. du Pont de Nemours, and provide an aes-thetically pleasing and comfortable outer surface for the sleeve 26. ~he sheets 52 may be attached to the sheets 36 and 38 by any -suitable mean~, such as by lines 54 of stitching along the side edges 40a and b and end edges 42a ~nd b which pass through tho sheets 52 and sheets 36 and 38 to secure the sheets togeth-r.
As shown in Fig. 2, the sheets 52 may have a plurality o~ open-lngs 56 to receive a plurality of connectors 58 which are se-cured to the sheet 36 and which communicate with the separate chambers in the sleeve 2$. As illustrated in Fig. 1, the con-nectors 58 are secured to the conduits 34a and b, such that the conduit~ separately communicate w~th chambers i~ the sleeve through the connectors 58.
2G As best shown in Figs. 2 and 3, the sleeves 26 may have a plurality of hook and loop strips 60 and 62, respectively, to releasably secure the sleeves a~out the patient's legs~ The hook strips 60 extend past one of the side edges 4Ob of the sleeve, while the loop st~ips 62 are secured to tha ootside of the outer sheet 52. During placement~ the sleeves 26 are wrapped around the patient's legs, and the ho~k strips 60 are.releasably attached to the associated loop str~p~ 62 on the outside of the 13~ 2 sleeves in order to hecure the sleeves on the legs and confine movement of the slee~es away from the patient's legs'when in-flated during operat~-on of the device.
As wlll b~'further discussed below, the controllers - ~2 of Figs. 9-11 int~rmittently form a plurality of fluid pre~-~ure pulse~ in a timed sequence during the periodic inflation or compresslon cycles, ln order to sequentlaIly lnitiate ln-flation of different chambors ln the ~leeve~. In the particular embodiments shown, the controllers 22 form three timed pressure pulses during each inflation cycle whlch are utilized to lnflate the six chambers in each of the sleeves, such that each pulse 18 associated with two chambers ln the sleeves. However, it will be understood that a timed pulse may be formed fo~ each of the chambers ln the sleeves, and that the number of timed pul~es may be varied in accordanco wlth the partl'cular type of sleeve belng u~ed ~n the devlce.
A graph of the prossures P formed ln the chambers of each sleeve with respect to time T 18 ~hown ln Flg. 8. The tlme to designatos the start of an lnflation cycle when a first pres-eure pulse 18 formed by the controller, and the flrst pulso ~
applied to the two lowermost chambers ln each of the sleeves at that time. As will be discussed below, the manifold separates the flrst pulse, an~ connects th- separated pulse~ to the two lowermost chambers 48a and ~8b, a~ designated on the corrospond-ing curves of Fig. 8. As shown, the pulse appl~ed to the lower-mo~t chamber 48a has a faster pre~sure ri~e time than the pul~oapplied to the ad~oinlng uppor chamber 48b, such that the rate of change of pressuro in the'lowerm~st chamber 48a is greater .

_9_ ;y ~

1 3~ 5'~

than the rate of change o~ pressure in the adjoining chamber 48b. Accordingly, the sleeve will exert a compressive pres-sure gradient against the limb which decreases from the lower-most chamber 48a to the ad~oining upp-r chamber 48b in the lower set of ad~olning chambers until the maximum pressura in the two chambers is reached and the chamber~ are filled.
The controller forms the second pres~ure ~ulse at the time tl during the inflation cycle, and inflation of the third and fourth chambers 48c and 48d in the sleeve is initi-ated at thi~ time. It will be seen that the device initiatesinflation of the third and fourth chambers while the first and ~econd chambers are still being filled from the flrst pres-sure pulse. The second pressure pul~e is al80 sep~rated by the manifold for the set of the third and fourth ad~oining~ch mber~
which have different pressure rise times, as shown, with the pressure rise time for the third chamber 48c being greater than the pressure rise time for the fourth chamber 48d. Thus, as in the case of the set of lowermost ad~oining chambers, the rate of pressure change in the third chamber 48c is greater than the rate of pre~sure change in the fourth chamber.48d, such that the set of intermediate ad~oining chambers al~o exerts a com-pressive pressure gradient against the limb which decrea~es from the third to fourth chamber. Additionally, it will be seen that the rates of pressure increases in the third and fourth chambers are less than those in the corresponding first and second chambers Accordingly, while the third and fourth chambers are being filled, -the pressures applied by the third and fourth chamber o the ~l-eve are 10~B than the pre--u~eo applied by the ~ t and -cond chambers, and the first, second, third, and fourth ch~m~ero thu~
exert a compressive pressure gradient which decreases from the .
13~

lowermost chamber 48a through the fourth chamber 48d.
At the ti~e t2 the controller $nitiates formation of the third pressure ~uIse for the fifth and sixth chambers 48e and 48f. As before, the pressure rise time in the fifth chamber 48e is greater than that in the uppermost sixth chamber 48f, such that the rate of change of pressure ln the f$fth chamber is greater than the rate of change of prossure in the ~l~th chamber.
~ccordingly, the set of ad~oining uppermost chambers applies a compress~ve pressure gradient against the patient' 8 limb which decreases from the fifth to sixth chambers. As shown, the pres-sure rise times in the fifth and sixth chambers are less than those in the four lowermost chambers, and while the fifth and sixth chambers are being filled, the pressure in the~e uppermo~t chambers is less than the pressures in the four lowermost cham-bers. Thus, the sleeve applies a compressive pressure gradientagainst the patient's limb which decrea~es from the lowermost chamber 48a to the uppermost chamber 48f in the sleeve. Once reached, the maximum pressures in the two lowermost chambers 48a and 48b are generally maintained throughout the inflation cycle while the remaining chambers are still being filled. Similarly, when the maximum pr~s6ures are attained in the third and fourth cbambers 48c and 48d, these pressures are generally maintained while the pressures are increased in the uppermost fifth and ~ixth chambers 48e and 48f. Maintenance of pre~sures ln a lower set of chambers may be sub~ect to slight diminution when inflation of an upper ~et of chambers 18 initiated. Finally, when the maximum pressures are obtained in the fifth and sixth chambers, all of the chamber~ have achieved ~heir maximum pressures during the inflation cycle. In a preferred ~orm, as shown, the maximum pressures attained in a lower set of chambers is greater than .. . .

13~ Z

those in an upper se~. of chambers, although the maximum pressures in the various ~et~ ~ay approach a compar~ble value, as desired.
In this manner, the ~evice intermittently applies n compressive pressuFe gradient by the sleeve during the inflation cycle~ whlch decreases from a lower part o~ the sleeve to an up~er part of the sleeve.
The controller initiates a deflation cycle at the time t3 when the air i8 released from the chambers, in order to deflate the chambers and release the precsures applied by the sieeves against the'limb.
The deflation cycle continues through a period of time until the Rubsequent'time to~ when the'controller again initiate~
formation of the first pressure pulse during a subsequent infla-tion cycle. The controller thu~ intermittently forms a plurality of pressure pulses in a timed sequonce for inflating the ~leeve~
during periodic inflation cycles, ~nd intermittently release~
pressure from the sleeves du~ing periodic deflation cycles between the inflatlon cycles.
As will be seen below, the time intervals between initia-tlon of the sequential pressure pulses, l.e., b-tween times to and tl, and between times tl and t2, is ad~u~table to modify the timed relationship of the pulse'se~uence. Additionally, the t$me inter-val elapsed during the inflation cycle, i.e.,~the time interval between times to and t3 is also ad~ustable to modify the duration of the periodic inflation cycles. Moreover, the time interval during the deflation cycles, i.e., the time interval between times t3 and to~ is adjustable to modify the duration of the periodic deflation cycles. Thus, the various time intervals associated with applying and removing the pressure gradients by the sleeves are suitably sdjustable accordiny to the physiology of the patient.

. 13~18~Z

The controller 22 and manifold 24 are illustrated in schematic form in Fig. 5. The controller 22 forms and applies the first pressure ~;ulse to a first manifold section 64a through the conduit 28a. The manifold section 64.a separates the first pulse through a pai~ of orifices 66a and 66b, and simultaneously .
fiupplies the separa~ed first pulses to separate manifold section~
68a and.68b, In tu~n, the manifola section 68a further separates the pul~e through orifices or ports 70a and 70b, which.permit free passage of gas therethrough or are of equal size, and simul-taneously suppl~es the separated pulses to the two lowermost cham-bers 48a.~in the pair of sleeves respectively through the associated .. conduits 34a and 34b; Similarly, the.manifold section 68b sepa-rates the pulse through ~imilar orifices or ports 70c and 70d, and simultaneou~ly ~upplies the separated p~l~e3 to the two second cham~er~ 48b ~n the pair of sleeve~ through the a~ociated conduit~
. 34a and 34b. As shown, the effective size of the orifice 66a is .. substantially greater than the effect~ve size of the orifice 66b in the manifold section 64a, such that the rate of flow of gas to the manifold section 68a is greater than the rate of flow of gas to the manifold section 68b. However, the effective sizes of the orifices 70a, b, c, and d in the sections 68a and b are such that the rate of gas flow through the section 68a to the two lowermo~t chambers 48a in the sleeve~ will be the 6ame, while the . rate of gas flow through the sect~on 68b to the two ~econd chambers 48b in the sleeves will also be the same although less than that to the two lowermost chambers.- Accordingly, the rate of gas flow through the section 64a to the two lowermost chambers 48a will be greater.than the rate of gas.flow through the section 64a to the two second chamber.s 48b, although the rate of flow to the two lowermost chambers 48a will be the ~ame and the rate of flow to the second chambers 48b will be the same. In this manner, the lowermo~t chambor~ aro filled at a greater rate than the ~econd chambers and have faster pressure rise times, such that a com-pre~sive pres~ure gradient is produc-d ln the first and s-cond chambers of the separate sle-ves wh~ch decroa~es from the fir~t chamber 48a to the second chamber 48b, The relative rate of gas flow through the manifold section 64a may be controlled by suit-able selection of the internal diameters of the orifices 66a and 66b.
The controller 22 forms and supplies the second pulse in the sequence to the manifold section 64b. The section-64b separates the second pulse through a pair of orificos 66c and 66d, with the orifice 66c ~aving an effective greater size than the orifice 66d, such that the resulting pulse supplied to the mani-fold section 68c will have a greater flow rate than the pulse sup-plied to the section 68d. As shown, the section 68c separates the pulse through orlfice~ 70e and 70f, an~ simultaneously supplies th~
separated pulses to the two third chambers 49c in the pair of sleeves through the associated conduits 34a and 34~. ~he effeo-tive sizes of the orifices 70e and f are such that the rate ofgas flow lnto the third chambers 48c of the two sleeves will be approximately the same. Similarly, the section 68d separates the pulse supplied to this section through orifices 70g and 70h, and simultaneously supplies the xesulting separated pulses to the two fourth chambers 48d of both sleeves through the agsociated con-duit 34a and 34b. Again, the effective sizes of the orifices 70g and 70h are ~uch that the rate of gas flow into the fourth chambers through conduit 34a and 34b will be approximately the same. However, since the effective size of orifice 65c is greater -~4-than that of orifice 66d,' the flow rate through section 68c to the third chambers 48c is greater than that through the sectlon 68d to the fourth chambers 48d ~hus, the pressure rlse times in the thlrd chambers of the sleeves 18 greater than'those ln tho fourth chambers of the sleeves, and the third and fourth chambers apply a compressivo pressuro gradlent ~gain~t th- patlont's llmb whioh decreases from the third to fourth chambers As previouffly di-cu~sea ln conneotion with Fig 8, the -cond pre-~ure pul-e $-formed by the controller 22 after formation of the first pulse, and the pressure rlse tlmes ln the chambers decreaso upwardly along the sleeve Accordingly, th- timed pulses supplied to th-lower four chambers in the sleeves result ln applicatlon of a com-pressive pressure aga~nst the patlent's limb which decreases from the lowermost chamber 48a to the fourth chamber 48d As will be discussed below, the controller 22 forms the second pressure pulse, which is supplied to the manlfold through tho conduit 28b, from the first pros~ure pul-e whlch 18 ~uppliod to the manlfold through tho conduit 28a ~ho controll-r form-the second pulso ln thls manner to produce ths progresslvely ~0 decreasing pressure rise times in the chamber sets and to prevent a possible inversion of the pressure gradients applied by the sleeves, since the second pressure pulse will not be formed unless the first pulse has been properly formed However~ since both manlfold sectlons 64a and b are supplied from tho first pulse after the second pul-o has been formed, a le~-er fllling pr-s-uro i~ avallable to the section 6ib than was initially available to the section 64a before formatlon .

of the second pulse. Thus, the effective size of the orifice 66c of ~ection 64b i8 made greater than that sf the corresponding orlfice 66a in the section 64a to obtain the desired compara~le, 0~ q,~although d~croa~ng, ~rpre~suro rl~- times in the correspond-lng flrst and third chambers. Similarly, the orlfice 66d of sec-tion 64b, although ~maller than the orifice 66c in the same sec-tion, has an effective greater slze than the corresponding orifice 66b in the section 64a to obtain the desired comparabie and decreasing pressure rise times in the corresponding second and fourth chambe~s. Thus, although the controller sup~lies gas for the second pressure pulse to the section 64b from the flrst pres-8ure pul~e, the effectlvely lncrea~ed orif~co ~zo~ in the sect~on 64b prov~de separate filling rates for the thlrd and fourth cham-bers which are comparable to, but preerably les~ than, the sepa-rat,e filling rates for the first and second chambers of the sleeve~
respectively, such that the pressure rise times in the third and fourth chambers are comparable to, but.preferably less than, the corresponding pressure rise times in the first and second chambers as previoufily discussed ln connection with Fi0. 8.
The controller then forms the third pui~e, snd sup-plies this p~lse to the manifold section 64c through the conduit 28c. ~he section 64~ separates the third pulse through flow con-, trol orifices 66e and 66f having effective different sizes, and simultaneously supplies thè separated pulses to the manifold sec-tions 68e and 68f. In turn, the sections 68e and f separate the pulses through orifices 70i, 70~, 70~, and 70 l, and simultane-ously ~upplies separated pulses to the fifth and ~ixth cham~er~
.

48e and 48f, respectively, of both sleeves through the associated conduits 34a and 34b. Accord$ngly, the rate of ga8 flow from the section 64c through orlfics 66e to the fifth chamber~ 48e is greator than that through the orlfice 66f to the uppermo~t sixth chamber~ 48f, such that the prossure rise times in the two fifth chambers of the sleeves is greater than that in the uppermost sixth chambers of the sleeve-. Thus, the fifth and sixth cham-bers apply a compressive pressure gradient against the patient's limb which decreases from the fifth to sixth chambers. Addi-tionally, since the third pressu~e pulse is delayed relative the first two pressure pul~es and ~ince the pressure rise time~ in the fifth and sixth chambers is less than the corresponding lower chambers, the pre~ures applied by the flfth and sixth cham-bor- aga~n~t the patient'~ llmb whil- being flllet are less than those applied by the lower four chamber~, as discussed in con-nection with Flg. 8, and the six chamber- of the two slooves thus comb$ne to apply a compressive pr---ure gradient against the limbs which decreases from the iowermo~t chnmbers 48a to the uppormost chambers 48f of the sleeves.
A- will be discussod below, tho third pros~ure pul~e supplied by the controller 22 through the conduit 28c i~ formed from the second pulse supplied through the conduit 28b in order to prevent an inversion of the desired pressure gradient and to provide the decreasing pressure rise times. Accordingly, the effective size of the orifice 66e in the section 64c is made greater than the offective size of the orifice 66c in th- section 64b, while the effective size of the orifice 66f in the section 64c is greater than the effective ~iz~ of the orifice 66d in the , section 64b, which '~180 permit~ the device to maintain the desired preRsurog ~n the lower chamber~ while filling tho upper-mo~t chambers. ~hus, although the lower four sleeve chambers are d'riven from the first and second pulses and the third pulse 5 is driven from the second pulse, the effective increased sizQ
of the orifice~ in the section 64c relative the sections 64b and 64a provides comparable, but decreased, pre~sure rise times in the uppermost fifth ~nd sixth chamber~, in a manner as previ-ously descr~bed.
Referring now to Figs. 5-7, the first, second, and third pressure pul~es are supplied to a manifold housing 72 through the conduits 28a, b, and c, respectively. ~he manner in wh~ch the first pre~sure pulse is separated by the manifold 24 for filling the first and second chamber~ 48a and 48b will 15 be described ln con~unction with Fig. 7.' The first pulse i8 supplied through the conduit 28a and inlet port 73 to a chan-nel 74 in the housing 72, ~nd the first pressure pu1se'is then separated through the orifices 66a and 66b in the housing 72.
As shown, the internal diameter of the orifice 66a i~ greater 20 than the interilal diameter of the orifice 66b, such that the rate of flow of gas from' the channel 74 into the housing chan-nel 76 is greater than the rate of flow from the channel 74 into the housing channel 78. The pul~e formed in the channel 76 i8 separated through orifice~ or outlet ports 7Oa and 7Ob having an 25 internal diameter of approximately the same size, or of suffi-ciently large size to prevent obstruction to passage therethrough, and the ~eparated pulses from orifices 70a and b are thon sepa-rately supplied to the two lowexmost chamber~ 48a of the pair of .

, ~3~

sleeves through the associ~ted conduits 34a and 34b. Similarly, the pulse formed in the channel 78 is ~eparated by the oxificeQ
or outlet ports 70c ~nd 7Od having an lnternal dlameter of approximate~y the same size as the orifices 70a and 70b or of non-obstructlve size. The separated pulse~ pass from the orlflce~ 70cand d through the as~ociated conduits 34a and b to the two second chambers 48b in the pair of sleeves.
In this manner, the first pulse passing th'roùgh the inlet port 73 is 3eparated lnto separate pulses in the channels 76 and 78, with the pul5e in the channel 76 having a faster pressure ri~e time than the pulse in the channel 78. In turn, the pul3e in the channel 76 i~ separated and supplied to the two lowermost chambers in the pair of sleeves, wh~le the pulse in the channel 78 is separated and supplied to the tw~ second channels in the pair of sleeves. Referring to Figs.'6 and 7, the second pressure pulse supplied to the manifold 24 through the conduit 28b is separated in a similar manner through a ser-ies of channels and orifices for filling the third and fourth chambers. Similarly, the third pulse, supplied to the'manifold 24 through the conduit 28c, is separated by interconnected chan-nel~ and orifices,' with the resultlng pulses being supplied to th~ uppermost flfth and sixth chambers. As shown, the manifold may have a pressure relief valve or pressure indicating device 81 secured to the housing 72 and communicating with the channel 74 or with any other channel or port, as desired.
In a preferred form, the controller 22 i8 composed of pneumatic component~, ~ince it is a preferred procedure to ~;3a~ z minimize electrical ~omponents in the potentially explosive environment of an operating room. Referring to Fig. 9, the controllex 22 has a regulator 100 connected to the source S
of pressurized gas in order to lower the supply pressure and dr~v- the oontroller clrcuitry. The regulator 100 i~ connect-ed to a two-pos~tlon switch 102 through a filter 104. When the switch 102 i8 placed in sn of~ cond$tion, the gas ~upply is removed from the circuitry components, while the swltch connects the supply to the component3 when placed in its on condition.
When the switch 102 is turned on, the air supply pas~ing through the switch 102 is connacted to port 105 of a two-posltlon or ~hlft valve 106. ~n a flrs~ conflguration of the valve, the ~upply i~ connected by the valve through the, valve port 108 to port 110 of shift valve 112, to port 114 of ohift valve 116, and to port 118 of a positive output timer 120. Actuation of the shift valve 112 at port 110 causes the valve 112 to connect its port 122 to valve port 124 and exhaust line 126. Similarly, actuation of the shift valve 116 at port 114 causes the valve 116 to connect its port 128 to port I30 and exhaust line 132. Al~o, the valve 106 connects the line 134 through its ports 136 and 138 to the exhaust line 140.
Accordlngly, when the shift valve 106 connects tho gas supply through its ports 105 and 108, the controller in-itiates a de~lation cycle during which gas passes from the sleeve chambers to the various exhaust lines, as will be seen below. At thiC5 time, the suppiy al~ ~nitiate~ the timer 120 ~3~ 2 which controls the d~ration of the deflation~cycle. The timer 120 is ad~ustable to modify the durat$on of the deflatlon cycle, and when the timer 120 times out, the timex actuates the shift valve 106 at port 142 to initiate an inflation cycle.
~he actuated valve 106 connects the gas supply through ports 105 and 136 to port 144 of a positive output timer 146, to port 148 of a positive output timer 150, to port 152 of a posi-tive output timer 154, and through the flow control valve 156 to o~9~ ."qTfport 158 of shift valve 116. ~he actuated t~T~e 106 also dis-connects $ts port 105 from port 108. ~he flow control valve 156 serves to reduce the relat~vely high préssure utilized to act-uate the pneumatic components of the circuitry to a lower pres-sure for inflating the chambers in the sleeves.
The gas supply passing through line 134 and valve 156 also passes through the conduit 28a to the manifold. Accordingly, the first pressure pulse is formed through the conduit 28a for f~lling the first and second chambers 48a and ~ of the sleevQs at this time. When the t$mer 154 times outO the gas supply i8 connected ~y the tlmer to port 160 of shift valve 116, which causes the valve 116 to connect its port 158 to port 128. Thus, the gas supply passing through flow contro'l valve 156 is connected through the shift valve 116 to the conduit 28b, and the second pre~sure pulse ~s formed and supplied to the manifold for inflat-ing the third and fourth chambers of the sleeves. It w~ll be ~een that the ~ontroll'er forms the second pressure pulse from the f$rst pre~ure pulse wh~ch is cont$nuou~1y upplled to th- manl-fold through the condu$t 28-. ~he time interval between $nlt$at$on ~3~ z of the first and secohd pressure pulses, respectively supplied through the conduits ~8a and 28b, is controlled by the ad~u~t-able timer 154. Acco~dingly, the duration between formation of the first and second pressure pulse~ may be modified by sim-ple adjustment of the timer l54.
When the timer 150 times out, the timer 150 connects the gas supply through the ~imer to port 162 of shift valve 112, causlng the vnlve to connect lts port 164 to port li2. ~he ga~
supply then passes through the ports 164 and 122 of shift valve 112 to the conduit 28c and manifold in order to inflate the fifth and sixth chambers of the ~leeves. Accordingly, the third pres-sure pulse supplied to the manifold i~ formed at this time by the control circuitry. It will be seen that the controller forms .
the th$rd pre~sure pul~e from the second pressure pulse supplied to conduit 28b, which in turn is formed from the first pr~ssure pulse, a~ previously described, and the first and second pressure pulse~ are continuously supplied to the manifold after the third pressure pul~e i~ pas~ed through conduit 28c. ~he time interval between initation of the second and third pulses i9 determined by the ad~ustable timer 150, and the timer 150 may be ad~usted to suitably modify the duration between the third pulse and the ear-lier pulses. Accordingly, the controller 22 $orm~ a timed ~e-quence of pressure pulse~, with the time intervals between the sequential pressure pulse~ being ad~ustable, as desired.
When the timer 146 times out, the timer 146 connects the gas supply through the timer to port 166 of shift valve 106.
At this tlme, th~ uhift va1ve 106 a~a1n connect~ ltu port 105 to !

Z

port 108, and disconn~cts the port 105 from port 136 of the valve, while the time~ 120 is again actuated to begin a de-flation cycle. It w$11 b~ seen that the t$mer 146 controls the duration of tbe $nflation cycl~es, since the deflat$on cycles are ln$tlated when the timer 146 tlmes out. The timer 146 al80 may be sultably ad~usted to modlfy the duration of the inflation cycles.
A~ prev~ou~ly dlscussed, when the de1at$on cycles are inltiated, the port 122 of shift valve 112 i8 connected to valve port 124 and the exhaust line 126. ThuJ, the two upper-most chambers 48e and 48f in the ~leeves are deflated through the conduit 28c and the exhaust line 126 at this t~me. Simllar-ly, when the valve li6,1s actuated at port 114, t~e port 128 of shift valve 116 i8 connected to valve port 130 and xhaust line 132, such that the third and fourth chambers 48c and 48d are - deflated through conduit 28b and the exhaust line 132. Finally, the shift valve 106 also connects its port 136 to port 138, such that the two lowermost chambers 48a and 48b are deflated through conduit 28a, valve ports 136 and 138, and exhau-t line 140. ~n thi~ manner,.the various chamber~ in th- sleeve- are d-flate~
during the deflation cycle. Referrlng to Fig. 5, ~t wlll be apparent that the pressure gradient, which decrease~ from a lower part o~ the sleeve to an upper part of the sleeve, is maintained during the deflation cycle, since the orifices in the sect$on 64c are effectively larger than th~ correspondlng orifices in the section 64b, while the orl~lce~ Jn the ~ection 64b are effeat-ively larger than the corre~pondln~ orlflae- ln tho sectlon 64a.

i :~3~ Z

~hus, the two uppermo~t chambers 48e and f deflate through the orifice~ 66e and 66f ~nd conduit 28c at a greater rate than the third and fourth Cham~ers 48c and d through the orifices 66c and 66d in section 64b and conduit 28b. Similarly, the third and fourth sleeve chambers deflate at a greater rate than tho two lowermost chambors 48a and b through orlfico- 6.6A and 66~ ln ~eot~on 64a and condult 28a. Aocordingly,.the compre-~lvo pressure grad~ent i~ malntalnsd durlng inflatlon and de1atlon ~f the sleeves.
~eferring again.to Flg. 9, it will ~e seen that the controller 22 intermittently forms the first, second, and third pressure pulses in a .timed sequence during perlod$c inflation or compre~sion cycies of the device. Also, the controller inter-mittently deflates the chambers ln the ~leeve during periodic deflation or decompre~sion cycles ~etween the per~odic lnflation cycles.
Another embodiment of the controller 22 of the present.
invention is illu~trated in Fig. 10. In thl~ embodimont, the source of pressurized gas S is connected to a regulator 200, a 20 filter 202, and an on-off switch 204, as described a~ove. When the switch 204 is placed in its off configuration, the gas supply S 1- removed from the pnoumatia component- of the controllor, whllo the ~upply S i- connected to the compononts when the switch i~ placed in lt~ on configuratlon.
. When the switch 204 18 turned on, tho ~r supply S 1 connected to port 206 o~ not gate 208. When pressure i~ ~bsent from port 210 of gate 208, the ~upply pa ~e~ through port 206 of . . . _ 1 3~J~

gate 208 to inlet ports 212 and 214 of ~ negative output timer 216. ~h~ supply actuato~ timer 216 at lt~ port 212, and the ~upply pa~ses througb port 214 of the timer to its outlet port 218. In tur~, the 8Upply i8 connected to port 220 of shift valve 222, to port 224 of not gate 226, to ports 228 and 230 of a positive output timer 232, and to ports 234 and 236 of a positive output timer 238. The pressure supply at port 224 of gate 226 prevents the gate 226 from connecting port 240 of the gate 226 to ports 242 and 244 of a negativo output timer 246.
The supply at valve port 220 actuates shift valve 222 which connects its port 248 to port 250, and thus the gas supply from switch 204 passes through the flow control valve 252, and ports 248 and 250 of sh~ft valve 222, to the conduit 28a and manifold. The flow control valve 252 reduces the rel-atively high pressure of the gas supply, which iB utilized to actuate the pneumatic component3 of the controller 22, to a lower pressure for inflation of the chambers in the sleeve.
~he conduit 28a ~ B connectod through tho manifol~ to tho two lowermost sleeve chambers 48a and b, as previously aescribed.
~hus, the device forms tbe first pressure pulse for filling the two lowermost cham~ers of the sleeves at the start of the inflation cycle.
When the positive output timeF 232 times out, the timer 232 connects the gas supply from its port 230 to port 256 of shift v~lve 258, which then connect~ its port 260 to port 262. Thus, the actuated valve 258 connocts th~ gas up-ply from the conduit 28a through i~8 ports 260 and 262 t~ the - ~ ~

13~ Z

conduit 28b and manifold for inflating the third and fourth OJ~ r~chambers 48~ and d of~the sleeves, and forms the second pres-sure pulse from the f~.rst pressure pulse at this time, with the time interval betIeen formation of the first and second 'pulses being controlled by the timer 232. As before, the duration between the first and second'pulses may be modified by suitable adjustment of the timer 232.
When the positive output timer 238 times out, the timer 238 connects the supply from its port 236 to port 264 of shift valve 266. The actuated valve 266 connects its port 268 to port 270, and thus connects the gas supply from conduit 28b through the valve ports 268 and 270 to the conduit 28c and manifold. Thus, the valve 266 forms the third pressure pulse from the second pulse at this time'for inflating the uppermost lS fifth and sixth chambers 48e and f in the sleevos. A~ ~efore, tho time ~nterval between the third pu180 and oarller pul~es controlled ~y the tlmer 238, and the duration'~etween the pulse~
may be modified ~y ~uitnble ad~ustment of the timer 238. It ~
noted at this time that the pneumatic components of the controller 22 are actuated by a portion of the circuitry which is separate from the gas supply passing through valve 252. and the conduits 28a, 28~, and 28c to the manifold and sleeves.
When the negative output timer 216 times out, the timer 216 removes the supply from port 220 of ~hift valve 222, from port 224 of gate 226, from ports 228 and 230 of timer 232, and from ports 234 and 236 of timer 238. The absence of pressure at port 224 of gate 226 causes the gate to pass the supply through gate ~L3~ 2 port 240 to ports 242 hnd 244 of the negative output timer 246 which initlates the st~rt of the deflation cycle. Conver~ely, the timer 216 initiat-s and controls the duration of tho infla-tion cycle, and the duration of the inflation and deflation cy-cles may be modifled ~y suita~le ad~ustment of the timers 216and 246, respectively.
When the timer 246 is actuated at its port 242, the timer 246 passes the gas supply from its port 244 to port 210 o gate 208, to port 274 of shift valve 222, to port 276 of shift valve 258, and to port 278 of shift valve 266. The pres-sure at port 210 of gate 208 cau~es the gste 208 to remove the supply from the port6 212 and 214 of the inflation timer 216.
At the same time, the pressure at port 274 of shift valve 222 actuate~ the v~lve whlch connects ~ts port 250 to port 280 and the exhaust line 282. Accordingly, the lowermost sleeve cham-bers 48a and b are connected by valve 222 to the exhaust line 282 through conduit 28~, and valve ports 250 and 280 of shift valve 222. Slmilarly, the pres~ure of port 276 of shift valve 258 actuates this valve whlch connect~ its port 262 to port 284 and the exhaust line 286. Thus, the thlrd and fourthlchambers 48c ani d of the sleeves are de~lated through conduit 2Bb, ports 262 and 284, and the exhaust line 286. Finally, the pre~sur- at valve port 278 actuates shift valve 266 which connects its port 270 to port 288 and the exhau~t line 290. Accordingly, th~ upper-most fifth and sixth chambers 48e and f of the sleeves are defla-ted through conduit 28c, valve ports 270 and 288 and the exhau-t line 290. It will be seen that all the chambers in the sloeve~
' . --27--13~

are simultaneously deflated through tha various exhaust lines 282, 286, and 290, and the compressive pressure gradient which .. . .
decreases from the lower to upper part of the sleeves is ma$n-tained during deflation of the s~eeves by the variously sized manifold orifices, in a manner as previou~ly described.
When the deflation t$mer 246 times out, the timer 246 removes the supply from port 210 of gate 208, as well as ports 274, 276, and 278 of valves 222, 258, and 266, re~pect-ively, and the gas supply is again connected from port 206 of gate 208 to ports 212 and 214 of timer 216 to initiate another inflation cycle. It will thus be seen that the controller 22 of Fiq. 10 also operates to intermittently form a plurality of pressure pulse~ in a timed sequence for inflating the ~leeves during periodic inflation cycles, and intermittently deflate lS the filled ~leeve cham~ers during periodic deflatlon cycles between the inflation cycles.
Another embodiment of the ~equential intermittent compression controller of the present invention is illustrated in Flg. 11. As before, the source S of pressurized gas is con-nected to a regulator 300, after which the source passes througha primary filter 302 and an oil filter 304 to a two-position - switch 306. Again, when the switch is plac d in its off condi-tion, the source or supply is removed from the pneumatic compon-ents of the circuitry, while the source is connected to th~ com-ponents when the switch 306 is placed in its on condition.
When the switch is turned on, the supply is connectedthrough the switch 306 to por~ 308 of shi~t valve 310. During 1 3~ 2 the deflation cycles, the valve 310 connects its port 308 to port 312, such that the gas supply is connected to port 314 of 8 positive output timer 316, to port 318 of 6hift valve 320, to po~t 322 of ~hlft valve 324, snd to port 326 of shift valve 328.
~he actuated shift valve 320 connects lts portP 330 to port 332 and exhaust 11ne 334, such that tho two lowermost chambers 48a and b of the sleeve~ are deflated through the man-ifold, the conduit 28a, the valve ports 330 and 332, and the ex-haust line 334. Also, the actuated shift valve 324 connects its port 336 to port 338 and the exhaust line 340. Accordingly, the valve 324 connect~ the third and fourt~ chambers 48c and d of the ~leeves through the manifold, the conduit 28b, the valve ports 336 and 338, and the exhaust line 340 in order to deflate the third and fourth chambers at this time. Finally, the actuated valve 328 c~nnects it~ port 342 to port 344 and the exhau~t line 346. The actuated valve 328 connects the two uppermo~t cham~ers 48e and f ~n the sleeves through the manifold, the conduit 28c, the valve ports 342 and 344, and the exhaust line 346 in order to deflate the fifth and sixth cham~ers of the sleeves. Accordingly, at the start of the deflation cycles the cham~ers in the sleeves are simultaneously deflated through the exhaust lines 334, 340, and 346.
- When the positive output timer 316 times out, the timer 316 connects the gas supply from port 312 of valve 310 through the timer 316 to port 350 of the shift valve 310 to actuate the valve at the start of an inflation cycle. ~he actuated valve 310 connects it~ port 308 to port 352 of the valve. ~n urn, the gas .
.

supply is connected to ~ort 354 of a positive output t$mer 356, to port 358 of a counte~ 360, to port 362 of shift valve 320, to port 364 of a positi'~e output timer 366, and to port 368 of a positive output timer 370. The actuated valve 320 connects its .port 372 to port 330, and, accordingly, the gas ~upply is connect-ed through the flow control valve 374, the valve ports 372 and 330, the conduit 28a, and the manifold to the two lowermo~t cham-bers 48a~and b of the sleeves. The flow control valve. j74 serves to reduce the relatively high pressure of the gas ~upply utilized to actuate the pneumatic.components of the oontroller circuitry, in ord~r to limit the supply pressure for inflating the sleeve~. -Accord~ngly, the ~ir~t pr-~sure pul~e i~ formed by the controller 22 at thi~ time to lnflate the first and second chambers in the sleeves. ' 15When the po~itive output timer 366 times out, the.ti-mer 366 connects the ga~ ~upply at port 364 of the t~mor to port 376 of shift'valve 32.4. The actuated shift'valve 324 connects its port 378 to.port 336 and the conduit 28b. Thus, the control-ler forms a second pressure pulse at this time from the first' pulse, with the second pulse being supplied through the conduit 28b and the manifold to the third and fourth chambers 48c and d in the~sleevQs. The interval of time between formation of the first and second pressure pulses i8 determined by the ad~ustable timer 366, and the duration between the pulses may be modified by su~table adjuctment of the timer 3q6.
When the po~itive output t~mer 370 time~ out, the timer 370 connects the supply through its port 368 to port 380 -.30-~ 3 ~ Z

of the shift valve 328. The actuated shift valve 328 connects its port 382 to port 342 and the conduit 28c. ~hus, the control-ler 22 forms the third pre~sure pul~o at this time which passes through tho conduit 28c and the manlfold to the uppermost cham-~ers 48e and f ln the sleevos, As beoro, the third pulse isformed ~rom the second pulse whlch 1~ suppl$ed throu~h the con-duit 28b. ~he interval of tlme between formatlon of the thlrd pulse and the earlier pulses is controlled by the timer 370, and the timer 370 may be suitably ad~usted to modify the duration between the pulses. Accordingly, the timed sequence of first, second, and third pulses may be modifled through adjustment of the tlmers 366 and 370.
~ he counter 360 1~ actuated at lts lnlet port 358 to ~ncrement the count~r 360 by one count corre~ponding to each ln-flat~on cycle of tho oontroller. A user of the device may thus determine the number of inflation cycles initiated by the device during use on a patient.
When the pos$tive output timer 356 times out, the timer 356 connects the gas supply through its port 354 to port 384 of sh$ft vnlve 310 to again ~tart a deflatlon cycle. As be-foro, the defl~tion tim-r 316 1~ actuated at port 314 whon tho shift valve 310 connects the supply through valve ports 308 and 312. Also, the actuated shift valves 320, 324, and 328 connect respective conduits 28a, 28b, and 28c to the exhaust lines 334, 340, and 346 to simultsneously deflate the chambers ln the sleeves while maintaining a graduated pressuro gradient, a~ prevlou~ly de-scribed. It will be seen that the ttmer 356 controls the duration 13~
.
, of the inflation cycles which may be sultably mod~fied by ad-~ustment of tho timer 3.~6. Accordingly, the controller 22 ln-termittently forms a pll~rallty of pressuro pulses ln a tlmed sequence during periodic lnflation cyclos, and the controller S intermittently de~lates the pres~urized chambers ~n the ~leeve~
dur~ng period~c deflation cycles which ta~e place between the inflation cycles.

~ - .
13~ 2 With reference to Figs. 12-20, full length ther-apeutic stockings and so-called tired-leg stockings of the type including elastomer-containing yarns which exert a compre~sive effe¢t on the leg portion covered by the ~tocking . 5 boot are well known. They have been con~tructed extending in lengths ranging from midthigh to the gluteal furrow in a great many constructions from a great variety of elastic fabric~.
They have, for instance, been made from powernet fabric, such as is described in U.S. Pat. No. 2,960,855, for example, cut to shapes resembling when relaxed the blanks of full-fashioned non-elastomeric stockings, being somewhat narrowed from such blanks. These powernet blanks and similarly ~haped knitted full-fashioned stocking blanks incorporating elastomeric yarns either in the knitted stitches or inlaid in non-elastomeric yarn stitches are generally seamed up the back with various loop, flatlock or overedge stitches to form finished stockings. Circu-lar knit stockings of non-elastomeric yarn jersey stitches with elastomer yarn inlaid therein are disclosed in the Herbert Knohl U.S. Pat. No. RE 25,046 originally issued Dec. 6, 1960. Other circular knit constructions including jersey knit course~ of elastomer-containing yarn alone and in combination with one or more course rounds of jersey stitches of non-elastomeric yarn are also well known, as are those with courses of jersey stitches and floats of elastomer-containing yarn alternating with jersey stitches of non-elastomeric yarn. Run-resistant elasti¢ fabric stockings have also been proposed.
With regard to the compro~siYe range of stocking~
presently marketed, the degree of compression exerted has ~een over a relatively large range. It is generally understood, how-ever, that in a properly fitted stocking the pres~ure should be ~L r.~g ~

greater at the ankle than at the stocking top whether the stocking be possessed of the relatively reduced compression typical of therapeutic stockings used in hospitals for the prophylactic treatment of the thromboembolic disease or of so-called tired-leg stockings or of the relatively higher compression typical of stockings recommended and used in the treatment of varicosities. In these stocking~ the pressure has been gradually reduced from the ankle to the stocking top or upper thigh when the stocking is properly fitted in order to increase the velocity of blood flow in the leg.
Full length stockings of the compressive type, regardless of the degree of compression exerted on the wearer's leg, have two problems. Because of extreme variation in the upper thigh dimensions of wearers even when other portions of the leg fall within a particular standard size range, full length stockings have been difficult to fit in the thigh area. A~ a result, manufacturers of non-custom stockings tend to make garments which will not bind the upper thigh~ of any significant proportion of wearers whose legs otherwise require a particular size stocking.
The tendency, then, is to make an enlarged thigh stocking whether it is enlarged by modifying a knit full-fashioned blank or a cut powernet blànk or is circular knit and enlarged by a wedge shaped insert. Such enlargement, however, sometimes causes the stocking to lose its self-support feature at the top. It i8 common prac-tice to make non-elastomeric stockings self supporting by attaching a thigh encircling garter band of elastic webbing whose leg-contacting inner surface is a non-s~ip material such as urethane elastomer. This bank may be attach~d under the stocking fabric but in most instances is attached i~ edge abutting relationship to the stocking welt, increasing the stocking length by the width -3~-of the band.
Full length stockings which have to be supported by an encircling garter band have had one undesirable feature, however. The elastic band, which i8 guite stiff and bears against the leg with some pressure, tends to irritate the upper inner thigh and to constrict the deep and superficial blood ve~el plexus there.
A desired stocking 1~ attained by sewing a band of the usual garterlike elastic webbing in edge abutting relation-ship to the stocking top welt with the band ends sewn to and separated by a wedge, fastened point downward, into a slit in the stocking upper thigh, the wedge top and the band top being aligned and forming the stocking top. The wedge which should be of a soft and readily conformable elastic fabric either in a single layer or a double layer, serves two functions. It reduces the binding in the upper thigh area when worn to a very small proportion o~ those who are otherwise fitted by a given size stocking, and it also constitutes the area which normally covers the deep and superficial blood vessel plexus in the upper inner thigh. The wedge may be centered in the area of the inner uppe~ thigh but this placement necessitates the manufacture of right and left leg stockings. Optionally, the wedge may be inserted centered over the front or bac~ fold line of the stock-ing from which position it may be rotated about a quarter turn to cover the inner thigh of either the right or left leg. The wedge sides are secured by sewing to the sides of the stocking slit and to the ends of the elastic webbing band by overedging or other appropriate stitching. ~ preferred wedge fabric is one in which elastomeric yarns are inlaid into jersey knit ~r~

stitches with the elastomeric yarns extending circumferentially when the wedge is secured in place. Double~fabric wedges are preferred, with the top folded edge forming a rolliresistant stoc~ing top in the wedge area. Doubled fabric preferably should be folded 80 that the normal outside surface is face to face and the normal inside surface forms the wedge's outer surface. Where a single thickness of fabric is utilized for the wedge, the top edge should be a soft selvage or it should be hemmed or overedged or sealed with a soft thermoplastic to make a ravel-resistant soft edge. A very effective wedge i8 in the form of an equilateral triangle about 6 inches plus the garter band width on each side.
Other variations may be made from circula~ knit stockings with finished or welted top ed~es and enlarged upper thigh portions by fastening to the inside of each a webbing band, corrugated slip-resistant surface exposed, and with a gap between the ends. The top edge of the band and stocking need not be but preferably are approximately aligned.
When an elastic biank is formed either by cutting from powernet or other suitable fabric or by full-fashioned knitting, the blank may be altered to include sufficient material in the upper thigh area to prevent binding. A projection above the top of the normal blank may be made in the area intended to cover the upper inner thigh or alternatively in the front center of the blank. If the top edge i5 to be double, the pro~ection should be double the width of the garter band; if single, the edge should be finished and the width should be the same as the garter band width. In this embodiment the partial circumference of elastic webbing band may be sewn ln edge to edge abutting relationship to the stocking top except in the area of the pro~ection with the bare corrugated ~lip-resistant surface of the band inside. Thereafter the ends of the band and the adjacent ends of the pro~ection either in ~ingle or doubled-S over form are sewn together, preferably in abutting relation-ship. A modification of this cut and ~ew or full-fashioned method involves sewing the band slip-resistant ~urface exposed inside a normal stocking top with a gap between the ends thereof, or if a single width pro~ection i present, folding down the pro~ection and sewing the projection ends to the ends of the band.
The stockings may be made adjustable in the upper thigh area by fastening means which permit that portion of the stocking top not containing the band of elastic webbing to be folded over the band and secured in place by a hook which is secured to the band and i8 caused to pierce the folded-over fabric. One or more separated hooks may be used but preferably a hook on either side is provided. ~he $abrlc is of 1008e enough construction as to permit piercing by the hook without injury.
Referring to the drawings in greater detail, Figs. 12 and 13, illustrate respectively the inner leg side and front view o~ a typical circular knit stocking 410 of the invention as worn, with a foot 470, an ankle 472, a boot 411, a knee 418, a thigh 419, and a soft readily conformable upper thigh insert 415 made of ~nitted elastic fabric The reciprocated heel 412 and the toe 413 are of typical heel and toe construction made from typical yarns preferably of stretch nylon. A partial round of elastic retention band 414 made with a corrugated slip-resistant inner surface of urethane elastomer is sewn to the ~L.3a~

upper narrow welt of the stocking proper projecting above the stocking welt 80 that its top forms a continuous line with the top of insert 415. The insert is overedged around its top edge and around its juncture with the slit stocking thigh 419 S and with band 414 by stitching 416. ~he insert 415 preferably - is symmetrical about the front or rear center line of the stocking so that it may be twisted in proper posltion to locate the insert 415 over the juncture "a" at the inner thigh, of the femoral, great saphenous, and superficial lateral cutaneous, pudendal and iliac veins. This plexus occurs approximately mid-way between the front of the thigh and the mid-inner thigh as depicted in FIG. 12.
FIG. 14 illustrates a preferred COnQtructiOn ln which the in~rt wedge 415a 18 of douSled fabric. The fold line 415b lS constitutes a portion of the top edge o~ the stocking. The fold is preferably made with the normal fabric face folded in face to face contact.
FIG. 13,15 and 16 illustrate the stocking of FIG. 12 with a hook 417 which is shown sewn to a portion of the band 414 adjacent the insert 415. The hook shown disengaged in FIGS. 13 and 15 is shown in FIG. 16 engaging a folded over portion of the top margin of the insert 415. This feature, which may be incor-porated on either or both sides of the insert 415, makes the upper thigh stocking portion adjustable in circumference.
FIG. 17 shows a typical fabric 420 suitable for the stockings of the invention, in which covered elastomeric yarns 421 are formed into courses of ~nitted jersey stitche~ alternat-ing with floats, the floats being across different wales in adjacent rounds. Yarns 422, which may be stretch synthetic yarns or usual nonstretch stocking yarns such as nylon, silk, ?~ z cotton, rayon,polypropylene and the like, are formed into jersey courses. The elastomeric yarn 423 is shown inlaid into one of the jersey courses of yarn 422.
FIG. 18 is the preferred typical fabric 430 suitable for the stockings of the invention. The yarn~ 432 are pre-ferably of Z-twist stretch nylon, while yarn~ 433 are pre-ferably of S-twist stretch nylon Sut may be any non-elastomeric yarn. A covered elastomeric yarn 431 is inlaid preferably into every other course as shown but optionally in every course of jersey stitches. I
FIG. 19 is another typical fabric 440 of the invention, in which covered elastomeric yarns 441 are formed into knit ed ~er~ey 6titches alternating with floats, the float~ being acro~
different wales in ad~acent rounds. Yarns 442, preferably non-elastomeric stocking yarns ~uch as synthetic or natural yarns including stretch synthetic yarns, are formed into course rounds of Jersey stitches.
FIG. 20 is another typica-l fabric 450, suitable for the stockings of this invention. The ~ersey knit fabric has alternating rows of stitches of synthetic or natural yarns 451 and covered elastomeric yarns 452.

- Using 10 filament stretch nylon 30/2 yarn, made up and knitted an automatic welt having a fully stretched circum-ference of 39 inches in the usual m~nner using a 401 needle Scott & Williams AMF 3 3/4 inches s~ocking knitting machine.
Immediately after the transfer, excpanged yarns to 70/1, 17 filament Z-twist nylon 66 yarn on ope feed and 70/1, 17 filament S-twist nylon 66 yarn on the other feed. Frame circumference fully stretched measured 42 inches. This frame was maintained to a point approximately at the upper calf, at which time the 1~3~ Z

frame was reduced abruptly because of machine-limitations but preferably within five to 10 courses to 32 inche~ fully stretched. This frame was maintained for approximately 120 course. Theframe thereafter was gradually reduced at a S constant rate by reducing stitch size until at the ankle the frame circumference measured 28 inches fully stretched.
Thereafter, for 150 cource rounds the frame remained at 28 i~ches circumference fully stretched. Thereafter the frame was gradually increased to the midpoint of the instep, at ` 10 which point the frame measured 32 inches in circumference fully stretched. Thereafter a reciprocated heel wa~ knitted in the usual manner. After completion of the heel, circular motion was resumed, the stitch being gradually reduced to a point between the heel and the toe to a circumference of 28 inches fully ~tretched. This circumference was maintained to the r~ng toe. Thereafter the ring toe including run-resist courses were knitted in the usual manner. Thereafter a reciprocated toe was knitted in the usual manner.
After the nylon frame was properly knitting, the elastomeric yarn was incorporated as follows: Immediately following completion of the top welt, the inlay feed was activated and a single covered elastomeric yarn having a 280 denier spandex core and a covering of 70/1, 34 filament stretch nylon 66 was inlaid in the course of jersey stitches knitted-off on the center feed. The elastomer should be metered in at a rate sufficient to produce a fabric having a fully stretched circumference of 38 inches. Knittin~ the frame incIuding the inlaid elastomeric yarn continued at that stretched circum-ference to a point just above the ca~.f, at which point the amou~t of metered elastomeric yarn was gradually reduced to the point at the upper calf where the circumference was 27 inches fully stretched. Th~ elastic yarn metering rate was maintained constant for about 100 nylon courses. Thereafter the elastomeric yarn was gradually increased per round to the midpoint of the instep, at which point the stocking had a fully stretched circumference of 26 inches. At that point the elastomeric yarn wa~ taken out and the reciprocated heel knitted. After completion of the heel, the elastomeric yarn was reintroduced in the usual manner and gradually decreased in amount per round to a point between the heel and toe, at which point the stocking foot fully strétched measured 22 inche~
in circumference. The elastomeric yarn was fed at this latter rate constantly for 60 course rounds, after which the elastomeric yarn rate was gradually increased to the ring toe, at which point lS the elastomeric yarn was taken out and the rlng toe including run-resist courses were knitted in the usual manner. Thereafter a reciprocated toe was knitted and the toe opening in the sole under the base of the toes was stitched clo~ed.
The finished stocking wa~ preboarded at 220F. to 230F. for 45 seconds; the total steam treatment and drying cycle lasted approximately 2 1/2 minutes; the total time, including build-up wa~ about 3 minutes. (Temperatures above 240F. are to be avoided if the most des1rable products are to be obtained.) Thereafter a l-inch wide band of typlcal elastic webbing of the type used for stocking garter tops was wrapped around the stocking, the corrugated slip-re~istant urethane elastomer side of the band outward. The ends of the band abutted at the front fold of the stocking but were not fastened together. The top of the stocking and the top of the ~band were sewn together in this posltion by overedge stitches.
The band was then folded upward so that it stood up from the stocking top, with the corrugated side in. A fa~tening hook, ~imilar to those illu~trated in FIG. 13, 15, and 16 was then sewn as illustrated to each side of the band about 1 1/2 inches from the band ends. (The distance may be anywhere from three-fourths inch to 2 inches.) The stocking was then slit between the band ends along the front fold for about 6 inches. A
diamond shaped piece of the same fabric as the stocking boot and about 6 inches across and 13 inches long, with the inlaid elastomeric yarn running t~ansversely, was folded to make sub-stantially equilateral triangle of doubled fabric. This triangular double-fabric wedge was inserted into the slit in the stocking, point downward, with its folded edge in alignment with the top of the band, and was overedge stitched, as is illu~trated in FIG. 14 to the side~ of the slit in the stocking and to the ends of the band of elastic webbing to complete the stocking.

13(:~9~Z

With regard to Fig. 21, according to the present invention, the elastic compression stockings 410 are first placed on the limbs, and the sequential compression device 20 or intermittent pneumatic compression ~IPC) is placed over the stockings 410, and both are used simultaneously on the limbs. A study was conducted to assess the prophy-lactic effectiveness for postoperative deep venous thrombosis (DVT) of a regimen employing the simultaneous use of graduated elastic compression stockings and IPC compared to a regimen of IPC alone.

PATIENTS AND METHODS

The study population was a consecutive group of patients undergoing general surgical procedures and without any condition which would make them unsuitable for scanning with the I-125 fibrinogen uptake test. Seventy-eight patients were studied, and all patients gave their informed consent for particip~ting in the study.
The mean age of the 43 male patients was 62.4 years ~+11.1 years), wh$1e the mean age of the 35 female patients was , 59.7 years ~l12.9 years).
The distribution of patients on the basis of age is illustrated in Table I. These groupings as a function of age are reflective of those typically encountered in this department and are consistent with those as reported in previous efficacy studies. A su~nary of operative procedures is given ln Table II.
Approximately one third of the patients were having operations for malignant disease and only seven of these procedures were considered palliative.

13~ 2 As the duration of anesthesia has been previously shown to be correlatable to risk for DVT development in the surgical patient, Table III provides a distribution of patients as a function of the duration of anesthesia. The mean duration of anesthesia was 72 minutes (+29 minutes).
In addition to venous stasis during anesthesia, po~toperative immobility i8 significant in the assessment of overall risk for DVT development. Immobility i8 defined as the number of days required before the patient is able to ambulate unaided.
The mean period of immobility for the study population was 64 hours (+14 hours).
On hospital admission, all patients were properly sized and fitted with graduated elastic compression stockings, - (Thigh ~ength TED, a trademark of The Xendall Company). At the time of surgery, one graduated compre~sion stoc~lng wa~
removed from either the right or the left leg (randomly allocated), and Intermittent Pne~matic Sequential Compression (sequential compression device SCD or IPC with full leg sleeves sold by The Kendall Company) was applied to both legs. The randomization of graduated compression stockings resulted in its application to right legs 51~ of the time and to left legs 49~ of the time.
The SCD remained until the patient was fully ambulant. At the time the SCD was discontinued, the previously removed graduated compression stocking was reapplied and all patients continued to wear graduated compression stockings on both extremities for the remainder of the study period.
All patients were assessed preoperatively for the presence of DVT using Doppler ultrasound and strain gauge plethysmography for maximum venous outflow measurements. Any positive sign for DVT preoperatively would exclude the patient from the study. Po~toperative diagnosis for DVT was conducted with dual screening approaches with any positive diagnosis being confirmed with contrast venography.

~;3C' t~9~3Z . ;

The I-125 fibrinogen uptake test was performed on the first, third, fifth, and seventh days after operation, using the Pitman isotope localization monitor. A sustained difference of more than 20 percent between adjacent sites on the leg was diagnostic of DVT, and the patient underwent contrast venography for clot confirmation.
To complement the I-125 fibrinogen uptake te~t, the patient also underwent Doppler ultrasound and maximum venou~ outflow asse~sment by strain gauge pl-thy~mography between the fLfth and seventh post-operative day. Again, a positive finding with either Doppler or strain gauge would be followed by contra~t venographic confirmation.

.

9~z R E S U L T S
"
The overall incidence of DVT in the patients studiQs was 9~. Results demonstrate complete Agreement between the I-125 dlagnosl~ and the contrast venography S diagnosls. In addit~on, there were no proxlmal vein thrombi detected. Thi~ finding is not surprising in view of the fact that full-leg compression prophylaxls was ln use across all of the patients ln the study, and the fact that the lntensive diagnostic survellence'allowed early treatment of anv confirmed distal DVT to prevent propagatlon. No pulmonary EMBOhII-wRre diagnosed ln any of the study patlent~.
Results from the alternate leg comparlson in Table IV
are categorized on a unilateral-bilateral basis. Of the untied cases, all 8iX of the DVT were found in the non-stocklng leg.
lS There was 0~ incldence of DVT ln the stocking leg and 7.7~
lncldence ln the non-stocking leg. The exact P-value for thls findlng is 0.0156 which 18 a statlstlcally slgnificant result.
The day of onset for thu DVT ls lndicated in Figure 1.
The only DVT diagnosis made d~ring the first postoperative day was the single bilateral case and reflects a patient with gastrlc carcinoma plus secondar$es. The remalnlng DVT, all unllateral in the non-~tocklng leg, were dlagnosed on days two through six.

Figure 22 i9 a Hlstogram showlng the day on which DVT was first detected in the stocking leg and the non-stocking leg.

13~

DISCUSSION

Only one patient in this series developed a deep venou~ thrombosi~ in a stocking leg, and this was associated with bllateral venous thrombosis. In all the other patlents studled the venou~ thrombosis occurred in the non-stoc~ing leg. The onset of venous thrombos~s occurs more commonly after the second po~toperative day, a findlng in keeplng wlth previous lnvestlgatlons. From this ~tudy lt would appear that the application of a graduated compression 6tocking beneath the ~eguentisl compression device is more effective in preventing deep venous thrombosis than SCD alone.
The methodology of this study does not exclude the possibility of an effect of one graduated elastic compression stocking on the blood flow in the opposite leg. However, results from similar risk patient populations relatlve to the demonstra-tion of efficacy of graduated compression in alternate leg series studies have been shown to be highly consistent with results demonstrating efficacy of graduated compression in alternate patient studles. ~n nssessment of this correlatlon demonstrates homogenelty ln trlal result~ relative to prophylactlc effective-ness. merefore, lt i9 argued that the design of this study to answer the fundamental queJtion of tQstlng the simutaneous use of graduated elastic compression and intermlttent pneumatic compression is appropriate.
The results are surprising as both compression modaliti~s have previously been thought to produce prophylactic effects in large part by reducing ~enous stasis via increasing linear blood flow velocity and dec~easing venous blood clearance time. In terms of llnear blood fl~w velocity, in was demonstrated t 1 3 ~

by Doppler ultrasound at the femoral vein in the inguinal ligament that graduated elastic compression stockings produced an average increase of 20~ over ba~e line. Using radiopaque dye methodology, an average increase wa~ previously measured in linear blood flow velocity of 33% over bàse line with graduated elastic compression stocklngs.
On the other hand, it was previously demonstrated that intermittent sequential pneumatic compression produces an increase of 240% in blood flow velocity as measured by Doppler ultrasound at the femoral vein. A comparison of these measurements by earlier investigators clearly demonstrates that intermittent sequential pneumatic compression increases linear blood velocity far in excess of that p~oduced by graduated elastic compression stockings. Therefore, if linear lS blood flow velocity were the only factor involved in the pro-phylactic mechanism for compression, then one would not expect the striking result relative to DVT formation demonstrated in this study. However, it was previously demonstrated that other hemodynamic parameters, namely venous capacitance and venous o~tflow, are significantly diminished from preoperative conditions to postoperative conditions.
While care must be taken in the interpretation of the3e data, the findings of this study in conjunction with the hemo-dynamic results from earlier investigations raise an important consideration of the mechanism by which graduated elastic compression stockings reduce DVT.
It was reported in animal model experiments a direct relationship between venodilation and endothelial damage, as well as, a direct relationship between venodilation and post-operative DVT in human subjects. The~e correlations and earliercanine model investigations provide a new appreciation of Vîrchow's Triad and in particular a new understanding of the interrelationship between venous stasis and vein wall injury.

It is suggested that venous stasis, as induced by surgery, can result in large changes in vein diameters with concomitant production of mechanical stress to induce microtears in the endothelial layer. The subsequent site of endothe~ial damage provides a location to potentiate the clotting process.
Durlng anesthesia and postoperative bed rest, venous dilation may occur as a result of muscle tone changes and con-comitant changes in venous capacitance. Intermittent pneumatic compression compresses the veins but for a relatively short period of time with a concomitant large increase blood flow velocity. It is possible that graduated compression stockings, while providing continuous stimulation of linear blood flow velocity also prevent dilation of the venous system in the lower extremities and reduce an additional aspect of thrombogenicity, namely the exposure of collagen.
This study indicates that the combined regimen of graduated elastic compression stockings with intermittent sequential pneumatic compression is a more effective prophylactic regimen than intermittent pneumatic compression alone.

-4~-. . ,''.,' " ,'. ' .

.
...... .. .
Years of k~e N~er of Pat~ents.
Male Female 40 - 50 ` 8 12 ` 51 - 60 . 5 6 .
71 - 80 .. 11; 9 Ov~r 80 , 0 2 .

.

.; :

.

- --5~--.. . .

13~9~2 . .
., ~ . _ .

.
.
, .
TA~E I~. OPER~CNS I~WED

.

.
C~ole~stect~qr . O 26 Colon Pesectlons 7 3 Sp~lenecta~y 4 2 Gastr~c Proeedures 3 - 5 I~pasotanies . 6 6 ~nc~s~onal Be~n~ , . 0 . 8 Other . 4 4 , .

, ~ , " ' ., . . , . . ... . _ . . , _ .. .. .

.

, .

N~ber of Patlent~ ~

.
I~ss than 1 ~ur ., 17 ;'' 1 to 2 2-ours ' SO

2 to 3 hour8 Greater t)~n 3 hours O
. . . . .

.:. . .

--52-- . ..

, _ .", ' ' ' - , ', '.
' .

.

Unilateral Bilateral tal Sto~c$ng ~ (78) 6~7.7%) 1 7~9%) Stoclc~ng Iegs (78~ 0~0%) , 1 1~1%) p - 0.0156 ~'s exz~ct test) ,:, ;

... . .

~; !

.;~ ",.~ ., , . , ; ,- . . , . y, , , ' . .. .

SUMMARY

The incidence of deep venous thrombosis (DVT) -was assessed in a series of seventy-eight patients under-going major surgical operations to compare the prophylactic effectiveness of intermittent sequential pneumatic compression alone to the simultaneous use of graduated compression stockings with intermittent sequential penumatic compression.
The diagnosis of DVT was determined with the I-125 fibrinogen uptake test, Doppler ultrasound, maximum venous outflow by strain gauge plethysmography and contrast venoqraphy. The incidence of DVT in non-stocking legs was 9% while that in the stocking legs was 1%, The simultaneous use of graduated elastic compression stockings with intermittent pneumatic compression (IPC) is more effective than IPC alone in the reduction of postoperative DVT.
The foregoing detailed description is given for clearness of understanding only, and no unnecessary limitations should be understood therefrom, as modifications will be obvious to those skilled in the art.

Claims (8)

1. A device for applying compressive pressures against a patient's limb, comprising in combination:
a stocking having a circumferentially elastic boot portion which applies a compressive pressure against the limb which decreases from the ankle to a top of the stocking;
an elongated pressure sleeve for enclosing a length of the patient's limb over the stocking, said sleeve having a plurality of separate fluid pressured chambers progressively arranged longitudinally along the sleeve from a lower portion of the limb to an upper portion of the limb proximal the patient's heart relative to said lower portion:
means for forming a plurality of fluid pressure pulses; and means for connecting the pressure pulses to chambers in the sleeve to apply a compressive pressure against the patient's limb by the sleeve which decreases from the lower to upper limb portions.

2. The device of claim 1 wherein the plurality of fluid pressure pulses are formed intermittently .

3. The device of claim 2 including means for forming pressure pulses in a timed sequence during periodic compression cycles.

4. The device of claim 3 wherein the pressure pulses are connected to separate chambers in the sleeve in an arrangement with later pulses in said sequence being connected to more upwardly located chambers in the sleeve.

5. The device of claim 1 or 2 including means for intermittently connecting the chambers to an exhaust means during periodic decompression cycles between said compression cycles.

6. The device of claim 1 or 2 wherein the stocking extends to a location adjacent the thigh region of the patient.

7. The device of claim 6 wherein the stocking has an elastic band having a pair of ends, with the ends of the band being separated by an area of soft conformable elastic fabric which extends downwardly for a substantial distance below the band for covering the plexus of deep and superficial blood vessels in the upper thigh of the patient.

8. The device of claim 7 wherein the soft conformable elastic fabric extends downwardly from the top of the stocking for a substantial distance below the band.