Treatment of pulmonary embolism

TREATMENT

OF PULMONARY

EMBOLISM*

MAJOR LEWIS S. PILCHER MEDICAL CORPS, ARMY OF THE UNITED STATES ULMONARY embohsm has aIways been one of the most tragic and most unpredictabIe of postoperative comphcations. In a recent comprehensive report from the Mayo CIinic’ covering over I 70,000 patients, puImonary emboli occurred in one out of every 250 surgica1 cases. In this group the extreme severity of this comphcation was emphasized by the fact that of the 897 patients who had pulmonary emboIi, 343 died. As usua1, many of these patients were heaIthy individuaIs, considered good risks, who had had eIective surgery and in whom such a compIication was entireIy unforeseen. In spite of the seriousness of postoperative puImonary emboIus, there has not been a cIear understanding of its exact etioIogy unti1 quite recentIy. Of course, it has Iong been recognized that the emboli originate from venous thrombi which break off into the Iarger vesseIs and are carried to the Iungs through the heart, but there has been considerabIe confusion as to the exact mechanism and site of the origina thrombus formation. It has graduaIIy been estabIished that there are two stages: First, the thrombosis of a periphera1 vein, often but not aIways associated with a variabIe degree of thrombophIebitis. This thrombosis then extends proximaIIy unti1 it reaches the junction with a Iarger vein. At this point the second stage starts, nameIy, the formation proximaIIy in the Iarger vein of a free propagating thrombus, with its base fixed to the thrombosis in the smaI1 vein and its tip floating free in the Iumen of the Iarger vein. It is this propagating thrombus and not the origina periphera1 thrombus which breaks off and forms the puImonary emboIus. The origina thrombus, especiaIIy

P

if there is any thrombophIebitis present, is usuaIIy IirmIy hxed to the vein waI1 and cannot readiIy break loose. It Iater may undergo organization and canaIization. In this connection, it is apparent that whereas a compIeteIy thrombosed vein, with or without an associated thrombophlebitis, wiI1 cause symptoms of venous obstruction periphera1 to the thrombosis, the reaIIy dangerous Iesion, the propagating thrombus, wiI1 cause onIy partia1 venous obstruction and may give sIight, if any, symptoms. It is aIso obvious that when a propagating thrombus breaks off there is no reason why another thrombus cannot readiIy reform and break off again, causing another emboIus, as is frequently known to occur. Furthermore, it is evident that whiIe a sIuggish circuIation favors the deveIopment of the primary periphera1 thrombosis and the secondary propagating thrombus, so an increased venous circuIation or vioIent motion of the vein waIIs wouId tend to break off the propagating thrombus. As attention has become focussed on the propagating or floating thrombus as the cause of puImonary emboIus it has become increasingIy apparent that a majority of these thrombi originate in the Legs, in the femora1 vein or its tributaries. Even in cases in which there has been evidence of IocaI thrombophebitis eIsewhere, as for example, cases of peIvic sepsis or appendicea1 abscess, postmortem examination has frequentIy shown that the source of the pulmonary embolus was from a thrombosis in the leg rather than from the area of IocaI sepsis. The actua1 incidence of venous thrombosis in the leg in patients who have had puImonary emboIi is aImost impossibIe to estimateaccurately,

* The patients in the cases reported in this paper were all treated in civilian practice at Newton, before the author entered the military service.

190

Massachusetts,

NEW SERIES VOL. LXIX,

No. 1

P&her-PuImonary

for in many cases the Iegs have not been specificaIIy examined for the possibility of venous disturbance, and even when examined the cIinica1 symptoms are often so sIight as to make the diagnosis questionable. The accumulating operative and postmortem evidence, however, indicates that a Iarge percentage, if not aImost aI1, puImonary emboIi originate from thrombosis occurring in the femoral venous system. The treatment of pulmonary emboIism has been approached in genera1 from three different angIes : First, treatment directed against the emboIi themselves after they have lodged in the Iungs; secondly, treatment of emboIi prophyIacticaIIy by preventing the formation of the origina peripheral thrombosis and the secondary thrombus; thirdIy, preventing the secondary, propagating thrombus from breaking off into the genera1 circuIation by mechanicaIIy occIuding the vein proxima1 to the thrombus. AS far as the first type of treatment goes, that is treatment of emboIi after they have lodged in the Iung, practicaIIy no important progress has been made, with two possibIe exceptions, nameIy, the prophyIactic use of the suIfonamides to prevent the deveIopment of a secondary pneumonia, and the use of papaverine to relieve spasm in the puImonary arteries (occasionaIIy a life-saving procedure). The dramatic operation advocated by TrendeIenburg, in which the embolus is removed from the pulmonary artery in a patient who is at the point of death, has proven impractica1 under the conditions which prevai1 in most hospitaIs in this country. The methods of prophyIactic treatment of emboIi aimed at preventing the occurrence of peripheral venous thrombosis, have chiefly been of two types: measures designed to combat sIuggish venous circuIation postoperatively; and the use of anticoaguIants to prevent intravascular cIotting. Among the measures advised, and in some few cIinics faithfully followed out, for stimuIating the periphera1 circulation,

Embolism

American Journal of Surgery

191

the most vaIuabIe and practica1 seem to have been the use of compression bandexercises and changes in ages, various position designed to minimize obstruction to return venous Aow, and the insistence on earIy and frequent motion of the Iegs. The use of heparin, and more recentIy of dicumaro1, to prevent intravascuIar cIotting has been haiIed by many as the fina solution of the treatment of puImonary emboIism, both before and after periphera1 venous thrombosis has occurred. The results so far have been promising. In the few clinics where heparin has been thoroughiy used, it has been demonstrated that the use of a carefuIIy controIIed continuous heparin drip wiI1, in many cases, prevent venous thrombosis, or at Ieast wiI1 prevent the formation of the propagating thrombus. Bauer,2 for exampIe, reported a group of fifty-three cases of femoral vein thrombosis; twenty-one of these were heparinized for ten days or longer and did not show any increased thrombosis or develop any embolic symptoms; thirty-two received no heparin and of these twentyfour showed extension of the femora1 thrombosis and three had massive pulmonary emboIi, of which two were fatal. In generaI, aIthough no deaths from puImonary embolism have been beIieved to occur during adequate heparinization, emboli and occasiona fataIities have occurred in patients treated with heparin after heparinization was stopped. This was presumably due to the formation of a propagating thrombus after the influence of the heparin had been removed. There ‘are aIso other drawbacks to the use of heparin. It is of very IittIe use, as Bauer points out, once the propagating thrombus is fully developed. Moreover, its whoIesaIe use prophyIacticaIIy in postoperative cases is for practical purposes out of the question due to the expense of the drug and the technica diffIcuIties of administering it continuously, in adequate dosage, whiIe at the same time avoiding overdosage with its danger of serious hemorrhage. These diffIcuIties have aIso prevented its use in

I92

American

Journal

of Surgery

Pilcher-Pulmonary

the average hospita1, many of which do not have the faciIities necessary to maintain adequate heparin technic. The discovery of dicumaro1, which is inexpensive and can be administered oraIIy, has overcome some of these objections to the whoIesaIe use of an anti-coaguIant drug prophyIacticaIIy. In 497 surgica1 cases recentIy reported from the Mayo Clinic3 in which dicumaro1 was given prophyIacticaIIy, the resuIts were extremeIy encouraging; ninety-one cases of postoperative puImonary emboIism or infarction were treated with dicumaro1 with no fatalities and onIy two recurrent emboIi, (in both of which cases the idea1 prothrombin IeveI had not been obtained with the drug treatment). Sixty-four patients with postoperative thrombophIebitis were simiIarIy treated with no fataIities and only two subsequent emboli; 302 surgica1 patients with no compIications were treated postoperatively with dicumaro1 without encountering any instances of thrombophIebitis or puImonary emboIism. The diffIcuIties encountered were chieff y in administering the dose of the drug so that an adequate elevation of prothrombin time was obtained without at the same time causing spontaneous bIeeding. Spontaneous bIeeding in this series occurred in forty-seven patients (IO per cent) and was cIassified as of moderate to severe degree in eighteen, with one fataIity from intestinaI hemorrhage. In a few cases20 it was not possible to obtain a satisfactory anticoaguIant effect with what was considered a safe dosage. In spite of its promising action, therefore, the use of dicumaro1 is stil1 associated with enough diffIcuIties and dangers to Iead the average surgeon to use it cautiousIy and in seIected cases. The Iast method of treating puImonary emboIi, namely, preventing a propagating thrombus which has aIready formed from breaking off into the genera1 circuIation by occluding the vein proximaIIy, is the method which up to the moment has appeaIed to us as the most practicat Iifesaving procedure, because of its simpIicity,

EmboIism

AUGUST, ,945

its safety, and the possibiIity of offering immediateIy to the patient and his famiIy positive assurance against recurrence or fatality. The operative procedure involved is of such a minor nature that it can be done safeIy even with criticaIIy iI patients. LocaI anesthesia is entireIy satisfactory. The usua1 technic is simpIe (assuming aIways an adequate knowledge of the regiona anatomy) : Locate the invoIved vein, open it, remove enough of the thrombus to estabIish how far it extends proximaIIy (it is not necessary to remove al1 the thrombus), and finally, Iigate the vein proximaIIy close to its junction with the next Iarger vein. Provided the ligation is cIose to the junction, any smaI1 floating thrombus which might form proximaIIy to the site of Iigation wouId be promptIy carried away by the swift ffow of bIood in the Iarger vein (usualIy a vein of the size of the iIiac or Iarger) before it couId grow to any dangerous size. This procedure has usuaIIy been appIied to the femora1 vein, because as noted above it is in this vein that most postoperative propagating thrombi are found to occur. However, occasionally thrombi may be found and removed in other veins, up to and in&ding the vena cava. The technic of approach to the femora1 vein need not be described except to mention that it is usually exposed at the upper end from the saphenous junction to beIow the profunda and that it is we11 to keep in mind that in this area the femora1 vein and artery are encIosed in a common sheath presenting a structure which when first exposed may appear to be the vein (or artery) alone. Care, of course, must be taken not to injure the femora1 nerve or its branches. A good description of the procedure in detai1 has been given by Homans who started advocating femoral ligation for pulmonary embolism in 1934. The diffIcuIty with this method of treatment is not, however, in the operative procedure, but in the diagnosis. To be most effective it should obviousIy be applied before the propagating thrombus

New SERIES VOL. LXIX,

No.

2

PiIcher-Pulmonary

has thrown off any emboli and that presents a considerable probIem, inasmuch as the majority of propagating thrombi cause shght if any cIinica1 symptoms until emboli are thrown off. Even when the presence of a thrombus is suspected, an accurate diagnosis of its presence, and even more of its location is difficult to make cIinicalIy. Although in many cases there may be suggestive signs, such as minor degrees of swelling or soreness of “Homans sign,” or one leg, a positive unexplained postoperative fever, these are rarely concIusive, but merely indicate the need of appIying some more definite diagnostic test. In the past various compIicated diagnostic procedures were suggested, but none were very satisfactory unti1 in 1940 Dougherty and Homans, foIlowing the work of DOS Santos,g in 1938, reported the use of venography in diagnosing a case of recurrent puImonary emboli which they found by this method were arising from a thrombus in the left femoral vein. At about the same time Bauer reported his more extensive experience with venography in the diagnosis of femoral thrombi in Sweden. At first it seemed that this procedure completely fiIIed the need for a diagnostic test by which the presence and exact Iocation of femora1 thrombi could be shown accurateIy and in time to be of vaIue, but further experience has shown that not even this diagnostic test is compIeteIy satisfactory, ahhough in most cases it is extremely helpful and in some instances concIusiveIy estabhshes the diagnosis. The question may stiI1 be raised by some whether femoral ligation should be advised in all cases when the presence of a propagating thrombus has been diagnosed. This question receives some backing from the fact that statistics indicate that onIy 20 per cent of patients who have had a propagating thrombus (as indicated by their having thrown off a smalI pulmonary embolus) subsequently succumb to a fata puImonary emboIism. The argument Ioses its appeaI, however, when the statistics

EmboIism

American Journal

ofsurgery

193

are read the other way and it is noted in the Mayo series, of the 343 patients who died from pulmonary emboIi, 33 per cent or I 14 were given warning of a fata emboIus by a preceding non-fata emboIus and could have undoubtedIy been saved if femora1 vein Iigation had been done routineIy after the first embolus. The technic and significance of venography has been thoroughIy viewed by DeBakey et al in a recent paper.* The exact procedure followed has varied with the individua1 surgeon and many of the technics reported seem unnecessarily complicated. Homans and Dougherty advise cutting down on the lesser saphenous vein behind the external maIIeolus and tying in a transfusion cannuIa. The dye is then injected through the cannula and washed in with a saIine infusion. Starr et a1.” use a simiIar technic. Linton advises injecting the superficial vein just inside the interna maI1eoIus and uses a bIood pressure cuff around the lower leg infIated to a pressure of 20 mm. of mercury, which forces the dye into the deep venous system. Both of these methods give exceIIent siIhouettes of the deep venous system but do not SimuItaneousIy fiI1 the superlicia1 system. The method which we have used with considerabIe satisfaction is to inject the dye into the superficia1 vein at the interna malIeoIus or into one of the other superficia1 veins of the foot. A size 19 needIe is used as in ordinary intravenous injections and usdaIly it is not necessary to cut down on the vein. No pain is caused if novocaine is ‘injected into the skin and if the needle is fuIIy in the vein. A bIood pressure cuff is pIaced around the upper, not the Iower Ieg, as high as possible up into the groin and inflated to 20 mm. of mercury at the time of the injection. By this method the dye is heId in both the superlicia1 and deep systems during the x-ray exposure, giving a comparative siIhouette of both systems simuItaneousIy. We have found that this comparative picture gives additiona1 vaIuable diagnostic information. In the norma venogram produced by this

‘94

American Journal

of Surgery

P&her-Pulmonary

method of injection three main venous trunks shouId be visibIe in the upper Ieg-the saphenous mediaIIy and the doubIe

A

EmboIism

Aucusl-,

19~5

determined onIy by comparing the venogram of the invoIved vein with a venogram of the same vein on the other (normaI) leg.

B

FIG. I. A, normal venogram: I, saphenous vein; 2, femoral vein; 3, profunda vein; 4, popIitea1 vein; 5, anterior and posterior tibia1 veins. B, normaI’venogram: I, saphenous vein; 2, femoral vein; 3, profunda vein; 4, popliteal vein; 5, anterior and posterior tibia1 veins.

shadow of femora1 and profunda centraIIy. (Fig. I.) The absence of one of these shadows is of great importance diagnosticaIIy. SimiIarIy in the Iower Ieg the three main systems of interna saphenous, anterior and posterior, tibia1 shouId normaIIy be present. It is of great importance to have venograms of both Iegs done with identica1 amounts of dye and identica1 technic, for in some instances when the ffoating portion of the thrombus onIy partiaIIy obstructs the vein, the vein shadow wiI1 not be absent but mereIy narrowed. Such narrowing can be definiteIy

This situation appIied to two of our cases in both of.which the diagnosis of a femora1 thrombus was confirmed at operation. AIthough we have found venography very vaIuabIe, a word of caution must be issued against its whoIesaIe, indiscriminate use. The dyes used are far from innocuous substances if used improperIy, as evidenced by the recent reports of occasiona fataIities from iodide poisoning foIlowing use of intravenoys dye in pyeIography. It is a wise precaution to test every patient for iodide sensitivity by having them hoId a few drops of the soIution on their tongue

NEW SERIES VOL. LXIX.

No.

z

P&her-Pulmonary

before using the soIution for intravenous injection. A second possibIe danger is the possibihty of the dye itseIf causing irritation of the vein when it is injected Ieading to new and extensive thrombosis in previously healthy veins. Rare instances of this complication having occurred have been reported. However, when an increased thrombosis foIIows an intravenous injection it is diffIcuIt to prove that the thrombosis wouId not have occurred anyway, even if the injection had not been given. In generaI, aIthough the danger must be kept in mind, it must aIso be remembered that the dyes used in venography have been used for many years in intravenous pyeIography without causing serious vein irritation. Starr,” in fact, points out that experimentaIIy the intima of the femoral vein has been exposed to the intravenous dyes used for as long as twenty-four hours “without gross evidence of injury”; and in tests which we conducted we found that a11 x-ray evidence of remaining dye in the veins had vanished within a few minutes after it had been injected. CASE

REPORTS

In the five cases reported beIow propagating femora1 thrombi were accurately diagnosed and successfuIIy treated by using the procedure of venography and femora1 ligation during a period of three months from the time our attention was first attracted to these methods of treatment. Further work was cut short by entry into military service. In three of the patients it was not until after the patient had suffered one or more pulmonary emboIi that the diagnosis was made, aIthough the diagnosis was suspected earlier in one and not seriousIy considered because of the minor nature of the symptoms. In the other two cases the diagnosis was made before there was any evidence of puImonary emboIism and prophyIactic femoraI vein expIoration was performed with remova of a femora1 thrombus. CASE

mitted

I. Mrs. E. J., age sixty-four, was adto the Newton Hospital, December 29,

EmboIism

Am&can

Journal dsurgery

194

194 I, for treatment of a compression fracture of the second Iumbar vertebra sustained two days previously when she had faIlen backwards into a bath tub. There was no evidence of injury to the legs and practicahy no externa1 evidence of trauma. The patient was placed on a hyperextension frame for five days. A posterior pIaster she11 was then apphed from hips to shouIders, and anchored in place with circuIar pIaster bandages across the peIvis and chest. The entire abdomen was left exposed because of a tremendous ventral hernia but the cast was steadied by pIacing a bed piIIow in the abdomina1 opening with a scultetus abdomina1 binder going around the cast and the pilIow. This apparatus immobiIized the spine satisfactorily and allowed the patient to’ turn freeIy in bed and exercise her legs. The patient’s hospita1 progress was uneventfu1 unti1 January rgth, sixteen days after appIication of the cast when she complained of pain in the Ieft upper quadrant. Her temperature rose to 100.4’~. for a few hours and then continued normal. There was no change in puIse or respirations. The white count rose to 13,900 with a norma differentia1. The pain was not very severe requiring no medication and there was no cough. At the time the probability of a smaII puImonary embolus was not suspected. Ten days later the patient compIained of simiIar pain in her right quadrant. In a few hours this became increasingIy severe and Iocaiized in the right Iower chest. Her temperatUre

rOSe

t0

102OF.,

p&2

t0

100,

reSpiI&OnS

to 30. The patient began to cough and raised bIoody sputum. The white count rose to 19,000 with a differential of 23 unsegmented polys, 66 segmented polys, IO Iymphocytes and I monocyte. The patient had had a sIight coId and a diagnosis of pneumonia was made which was confirmed when an x-ray showed diffuse, consoIidation of the entire right lower lobe. The sputum, however, showed very few pneumococci. In accordance with the diagnosis of pneumonia the patient was pIaced on fuI1 doses of suIfadiazine but in spite of a bIood IeveI of 12.2 mg. per cent, the temperature, pulse, and respirations remained unchanged for five days and the cough and bIoody sputum were unchanged. When the pain in the right chest, occasional bloody sputum and Iow grade temperature continued for ten more days we were

196

American

Journal

of Surgery

P&her-Pulmonary

convinced that we had been dealing with a septic pulmonary embolus rather than a true

LEFT

LEFT

Embolism

Aucusr,

,945

patient was aIlowed to move more freeIy in bed. Five days Iater, she suddenly had a severe

RIGHT

RIGHT

FIG. 2. A. Case I. Mrs. E. J. Left, venogram of right leg, normaI; rigbf, venogram of Ieft kg; saphenous vein only Wed. B, Case venogram of right leg, normal; right, venogram of Ieft I. Left, kg; saphenous vein onIy fiIIed.

pneumonia. The difficulty was that there was no obvious source for an embolism. On February I6th, three weeks after this episode the body cast was removed and the

pain in her Ieft chest with acute dyspnea. Her p&e rose to 160, respirations to 45. There couId be no doubt about the diagnosis of a typica severe pulmonary embolus. As is com-

NEW SERIES VOL.LXIX.

No. z

P&her-PuImonary

mon in acute emboIus, x-ray did not immediateIy demonstrate the infarct in the Ieft Iung although it outIined cIearIy the residua1 healing infarct in the right Iung. An x-ray taken one week Iater cIearIy showed the new infarct. Again the patient’s Iegs were carefuIIy examined but showed absoIuteIy no edema or tenderness. However, on questioning a story was eIicited of severa transitory attacks of fever, redness and swelling of the left leg occuring at intervals during the past ten years, suggesting a recurrent phIebitis of the Ieft Ieg. With this Iead, bilateral venograms were immediateIy done, injecting 20 cc. of diodrast in the dorsal veins of each foot. A compIete bIock of the left popIitea1 and femora1 vein was cIearIy demonstrated while at the same time contro1 venograms of the right leg showed good fiIIing of the femora1 and saphenous systems. (Fig. 2.) The Ieft saphenous vein was shown to be patent. On this basis ligation of the Ieft femoral vein beIow its juncture with the saphenous was advised. A few hours Iater the Ieft femora1 vein was expIored through a vertica1 incision extending for 3 inches dista1 to the saphenous opening. Local anesthesia was used. On opening the vein it was found to be compIeteIy occIuded by a soft cIot. There was IittIe evidence of inff ammation of the vein waI1. The thrombus was removed proximaIIy up to the saphenous juncture where free circulation was demonstrated. The femoral vein was then Iigated cIose to the saphenous and a short segment of the vein excised. This segment together with part of the thrombus was saved for examination. The report of the pathoIogica1 examination of these specimens was as follows: “Specimen (1) consists of 4 to 3 mm. segment of femoral vein which has a thin wall. The Iumen is empty. (2) A smaI1 irreguIar fragment of reddish and pinkish gray thrombotic matter which appears to have come from the vein and measures 0.8 mm. long and 0.6 cm. in diameter. Microscopical examination of section through the femoral vein shows slight irreguIar thickening of the waI1. There is a smaI1 patchy 1ymphocytic infiItration present. Section through the thrombotic matter shows it to be of variegated type. “Diagnosis: (I) Segment femora1 vein. SIight chronic phiebitis. (2) Variegated thrombus.” (Such Ioose thrombi in veins showing IittIe evidence of phlebitis are typica of the condition found in propagating thrombi.)

EmboIism

American Journal of Surgery

I97

FoIIowing this procedure the patient deveIoped sIight edema of the Ieft Ieg which cIeared up completely in four weeks. The most recent puImonary infarct in the Ieft chest cIeared sIowIy and for this reason the patient was not aIIowed out of bed unti1 three weeks She was discharged to a after the ligation. nursing home three days later. At the Iast examination, six months after the operation she had returned to her home in Connecticut and was living a normal Iife with no discomfort or edema in either Ieg. CASE II. Mr. P. Z., age fifty-eight was admitted to the Newton HospitaI, February 7, 1942, for treatment of swelling and redness of the right Iower Ieg. Two weeks previous!y he had sIipped and faIlen on the stairs while working. At this time he felt a pain in his right caIf. Two days Iater his foot began to sweI1 and then the entire Iower leg became markedly reddened and swoIIen to the knee. He had no chihs or fever and Iittle discomfort. The sweIIing and redness improved with ten days’ rest and eIevation at home but as soon as he tried walking again the foot became swollen and was cyanotic when kept dependent for any Iength of time. He was sent into the hospital for paravertebra1 bIock with a diagnosis of traumatic thrombophIebitis, right popIitea1 vein. On admission to the hospita1 his genera1 examination was essentially normal, blood urine and bIood studies pressure I20/75, were normal. LocaI examination showed the right Iower Ieg reddened, tense and swoIIen with sIight tenderness over the calf only. The superficial veins beIow the knee were obviousIy dilated in comparison with the Ieft Ieg. TweIve hours after admission to the hospital, before any treatment had been given, he awoke in the night with severe pain in the left chest and acute dyspnea. His puIse was 120, respirations labored. The attack subsided in two hours and the patient feIt we11 the next morning, but showed a few rlIes at the Ieft base which had not been there twenty-four hours before. Immediate x-ray showed no consohdation, but a Iater x-ray showed a heaIing infarct at the left base. (The lung changes, which foIIow an embolus and produce an infarct do not deveIop fuIIy for several hours or days and obviousIy x-ray changes are not demonstrabIe for a simiIar period of time.) EIectrocardiographic studies were essentiaIIy

198

American

Journalof

Surgery

P&her-Pdmonary

negative, showing sIight Ieft ventricmar preponderance but no evidence suggestive of coronary disease or a Iarge pulmonary emboIus. AIthough it was beheved definiteIy that this patient had had a smaI1 puImonary emboIus our first reaction was to foIIow the oId Iaissezfaire schoo1 and take our chances on the future. LocaI treatment-bed rest, heat and eIevation -was supplemented with two paravertebra1 sympathetic blocks but after two weeks the right leg stiI1 became swoIIen and cyanotic whenever it was hung down for even a few minutes, and the diIatation of the superficia1 veins persisted even with the leg in bed. This seemed to indicate a fairly extensive thrombosis of the deep venous system probabIy extending up the femora1 vein to the saphenous. At this time, therefore, it was decided to expIore and Iigate the femora1 vein for the dual of preventing further puImonary purpose emboIi and restoring normal circuIation to the Iower leg (which wouId be accompIished by interruption of the sympathetic innervation of the femoral vein attendant upon Iigation and excision of a segment of the vein). On February zest, two weeks after admission, operation was performed, exposing the right femoral vein for a distance of three inches below its junction with the saphenous vein. A vertical incision was used. A firm partiaIIy organized thrombus was found IiIIing the femora1 vein up to but not beyond the profunda branch. Norma1 circulation was demonstrated in the profunda and saphenous veins. The thrombus was removed proximaIIy up to the junction with the profunda vein and the femora1 vein was then ligated just beIow its junction with the profunda. A half inch segment of femora1 vein showing considerable evidence of periphIebitis was removed. (It was this periphIebitis, of course, rather than the thrombus inside the vein which was responsibIe for the edema, etc., of the Iower Ieg, for as noted above the thrombus alone may be present without causing any definite periphera1 edema.) FoIIowing this operation the redness and edema in the lower Ieg rapidIy subsided. A venogram was done ten days postoperativeIy and showed good compensatory circulation through the saphenous vein. Two weeks after operation the patient was walking about freely with absoIutely no edema or cyanosis and went home one week Iater ready to return to work. The prompt reIief of the edema and cyanosis was particuIariIy striking since this patient

EmboIism

AUGUST, ,945

had had four weeks of conservative treatment in bed before femoral Iigation without relief of the edema. CASE III. Mrs. E. D., age fifty-seven, was admitted to the Newton Hospital ApriI I, 1942, for treatment of an acute pulmonary emboIus in the Ieft lung. On December 30, 1941, three months previously, she had been hit by an automobile sustaining a compound, badly comminuted fracture of the Ieft tibia and fibuIa with destruction of both Iower tabIes of the knee joint. There was extensive subcutaneous tissue damage with considerabIe trauma to the veins of the anterior and medial aspects of the lower leg in the region just below the knee. An open reduction was performed with satisfactory re-approximation and wiring of the comminuted fragments of the knee joint. A circular pIaster cast was then appIied extending up to the upper thigh. There was no infection of the wound and the patient ran a smooth course foIIowing the operation except for a low grade temperature just over 99’F. which persisted for over two weeks. AIso, on the tenth postoperative day the patient suddenly developed sIight swelling of her thigh inside the cast, so that her cast suddenly feIt tight. The wound was entirely cIean and there was no evidence of deep infection such as swolIen inguina1 nodes. The possibiIity that the sweIIing might have been caused by a miId thrombophIebitis of the popIitea1 and femora1 veins was considered, but the sweIIing subsided in a few days and it was beIieved that no further treatment was indicated. The question of doing a prophylactic Iigation of the femora1 vein to guard against the danger of a pulmonary emboIus was raised, but because of the minor nature of the patient’s symptoms it was voted down. When two and a haIf months Iater the patient left the hospita1 walking on crutches the incident had been practicaIIy forgotten. When the patient got home she began to receive more vigorous physiotherapy than she had been getting in the hospita1. After a few days she complained that this caused some soreness in her knee and Iower Ieg so that orders were given to have it administered Iess violently. On this regime the soreness subsided and the patient began to walk with a brace without her crutches. Then suddenIy one night ten days after Ieaving the hospital and three months after her operation, she awoke with severe pain in the Ieft chest, cough and acute

NEW SERIES VOL. LXIX,

No.

2

Pdcher-Pdmonary

dyspnea-the typical picture of a puImonary embolus. On admission to the hospita1 the patient was critically iII and for the first twenty-four hours

Embolism

American

~~~~~~~

of surgery

Ieft femoraI vein did not fiI1 in the venogram. The left profunda and the Ieft saphenous were Iarger than normal, a situation which had been noted in a previous case (Case IV reported

FIG. 3. Case III. A, venogram of left Ieg; popliteal and anterior tibia1 veins not filled. venogram of left leg; popliteal and anterior tibia1 veins not filIed.

was placed in an oxygen tent. The temperature rose to 103.4'~. in twenty-four hours but further sepsis was controIIed by the use of fuI1 doses of suIfadiazine oraIIy. X-ray studies twenty-four hours after admission and at Iater dates confirmed the presence of a Iarge area of puImonary infarction in the Iower left Iung. On the basis of the history it was believed certain that the origin of the embolus was a propagating thrombus of the Ieft femoral vein probably originating from a traumatic thrombophIebitis of the anterior tibia1 venous system. BiIateraI venograms were done which confirmed this diagnosis by demonstrating Iack of fiIIing of the left anterior tibia1 system. (Fig. 3.)The contro1 venogram showed norma fiIling of the right anterior tibia1 system. The

199

B,

below) in which a propagating thrombus had been removed from the femora1 vein. Because of the patient’s critica condition exploration of the Ieft femora1 vein was postponed for three days. On April 4th, however, the Ieft femora1 vein was exposed under Iocal anesthesia through the usua1 vertical incision. On opening the vein close to its juncture with the saphenous no thrombus was found but it was demonstrated that while there was a free retrograde flow of blood from the iliac vein back into the femoral there was no flow of bIood from below, indicating a compIete bIock of the Iower end of the femoral system. It was beheved that originally the propagating thrombus had probably occupied the entire femora1 vein but that the portion occupying the upper

200

American

Journal

of Surgery

P&her-PuImonary

femoraI had broken off to produce the recent emboIus and had not yet reformed. Further exploration was not done but the femoral vein

EmboIism

AUGUST,

,945

was sent home walking (with crutches) and with Iess edema of her foot than she had had at any time since she had been out of bed after

FIG. 4. Case IV. A, venogram shows singIe narrow trunk instead of broad doubIe shadow of femora1 and profunda veins normaIIy seen. n, venogram shows single narrow trunk instead of broad double shadow of femoral and orofunda veins normaIIy seen. Note abnormal collateral veins. I

was ligated just beIow the profunda, removing a smaI1 haIf inch segment of the vein. (In performing ,this operation we remove a segment of the vein routineIy, beIieving that such removal with its interruption of sympathetic fibers relieves the edema of periphera1 thrombophIebitis, and wiI1 also act as insurance against the formation of edema which might resuIt from periphlebitis foIlowing expIoration of the femoral vein. In this case there was some edema of the foot present before operation and this had almost entireIy disappeared within the next two weeks.) Because of the Iarge pulmonary infarct this patient was not aIlowed out of bed for two weeks after the operation. One week Iater she

her accident. During the subsequent five months the patient continued in good heaIth and graduaIIy regained good function of her knee, aIthough the enforced bed rest necessitated by the puImonary infarction slowed up the motion of her knee considerabIy. In this connection it is interesting to speculate how much better off the patient would have been in every respect if prophyIactic femora1 Iigation had been done when it was first considered at the time when the patient first showed suggestive signs of femora1 thrombosis, ten days after her injury. CASE IV. Mrs. E. M., age fifty-two, was admitted to the Newton HospitaI, February 17 1942, for repair of a huge ventra1 hernia in an

NEW SERIES VOL. LXIX,

No.

2

P&her-PuImonary

oId choIecystectomy scar. In her past history she had had four Iaparotomies, two of which were cesareans, one appendectomy and finaIIy the choiecystectomy, done ten years ago. She had noticed varicose veins in both Iegs but never had a milk Ieg or any symptoms suggestive of phlebitis. Physical examination on admission was negative except for the Iarge ventra1 hernia, marked genera1 obesity, and moderate varicose veins of both Iegs. On February rgth, under spinal anesthesia the ventra1 hernia was repaired. Because of the Iarge size of the hernia defect and the diffrcuIty in approximation, a tight scultetus binder was ordered and kept tightly in place for three weeks. Postoperatively the patient ran a febriIe course with temperature between IOO and 101”~. daiIy. This was accounted for by a cystitis which developed from an inIying catheter and from a stitch abscess. The stitch abscess drained on the tenth day but continued to discharge pus for two weeks. On March 5th, on her fourteenth postoperative day, she suddenIy compIained of pain in her Ieft calf. There was very sIight tenderness of the caIf with a positive Homan’s sign but no edema or redness and no increased fever or chiIIs. BiIateraI venograms were immediateIy done using IO cc. of diodrast in each leg. The Ieft popliteal vein did not Ii11 and the left femoraI was narrowed whiIe the left saphenous was Iarger than normal. (Fig. 4.) This indicated to us a thrombophIebitis of the left popliteal vein with a propagating thrombus in the Ieft femora1 vein. With this definite evidence of thrombosis in the Ieg it was decided to do an immediate prophylactic Iigation of the Ieft femoral vein. ApproximateIy twelve hours after the onset of the leg pain the Ieft femora1 vein was expIored through a vertical incision extending 3 inches distally from the saphenous opening. The vein was opened and the venous flow from below found to be completely obstructed. There was a considerabIe amount of soft cIot which was removed with suction. The absence of periphIebitis and the presence of a soft thrombus which was attached to the vein wall confirmed the preoperative diagnosis of propagating thrombus of the femora1 vein. A half-inch segment of the vein with attached clot was excised and the ends of the vein Iigated. The pathoIogica1 report of the specimen of vein and thrombus removed was as fohows;

EmboIism

American journal or surgery

201

“Specimen consists of a short resected segment of vein 0.7 cm. Iong and 0.5 cm. in diameter. The Iumen is occIuded by thrombotic material. Microscopica examination of section through the segment of femora1 vein shows the lumen to be occIuded by variegated thrombotic material. The wall presents moderate irreguIar fibrous internal thickening and shows a Iight diffuse infiltration of lymphocytes and a few neutrophiIs. “ Diagnosis : Femoral vein. Thrombophlebitis.” FolIowing the operation there was sIight edema of the Iower leg for ten days. Most of this edema seemed to be centered over an area on the dorsum of the foot where a smaI1 amount of diodrast had been injected outside the vein and it was believed that the edema was probabIy due to this rather than to the Iigation of the femoral vein. The patient was aIIowed out of bed on the eleventh day after ligation and went home one week Iater. The edema of the Ieg had aImost disappeared except for the IocaI area on the dorsum of the foot when she Ieft the hospita1 and had entireIy cIeared up one month Iater. CASE v. Mr. W. H., age sixty-eight, was admitted to the WaItham Hospital, March 8, 1942, with a general peritonitis due to a ruptured appendix of several hours’ duration. His past history was negative except for a biIatera1 hernia repair twenty years ago. Shortly after admission appendectomy with drainage was performed and 12 Gm. of suIfaniIamide crystals introduced into the peritoneum. Sulfadiazine was given by mouth in fuI1 doses for the next two weeks. He had a very stormy postoperative course marked by almost continua1 deIirium, recurrent vomiting and several episodes of bIeeding from his wound, cuIminating in complete evisceration of the incision on the twenty-second postoperative day. This was resutured. FolIowing resuture (and the omission of the sulfadiazine) he ran a smooth postoperative course and was up and walking ten days after this secondary operation. Five days later he suddenIy noticed that his right ankle and leg became swoIIen and cyanotic when he put his legs down. A clinica diagnosis of probabIe thrombosis of the right femora1 vein was made with a possibility (in view of his recent peIvic sepsis) that the thrombus might extend even higher into the right iliac vesseIs. A biIatera1 venogram was done and settIed the diagnosis definitely by reveaIing

202

American Journal of Surgery

P&her-PuImonary

a compIete bIock of the right femoral vein with a patent right saphenous vein (indicating that the bIock did not extend above the junction of the saphenous with the femoral vein). With this definite diagnosis and in view of the patient’s postoperative course it was beheved that this patient had a better than average chance of deveIoping a puImonary emboIus. Accordingly expIoration of the right femoral vein was advised. On ApriI 18th, under novocaine anesthesia the upper end of the vein was exposed through the usua1 vertica1 incision. The vein was opened and found to be occIuded up to but not beyond the saphenous opening by a firm organized thrombus IightIy attached to the vein waI1. After this thrombus was removed the iliac vein was thoroughly suctioned and no further cIot found. The femora1 vein was then Iigated just distal to the saphenous vein and a haIf inch segment of the vein excised. The specimens were sent to the pathoIogist for examination and the report returned, “Organizing thrombus of femora1 vein with sIight thrombophlebitis of vein waI1.” FoIlowing this femoral Iigation the patient made an exceIIent recovery. He was aIIowed out of bed two days after the operation and sent home walking freely and with no edema of his foot seven days Iater. In this case, as in Case IV, femoral exploration and remova of the threatening thrombus was done entireIy prophyIacticaIIy as soon as the diagnosis was made and before the patient had a chance to deveIop a pulmonary embolus. In neither case was the patient bothered by the procedure and the hospital stay was shortened rather than Iengthened due to the prompt control of the thrombophIebitic edema which folIowed the partia1 resection of the upper femora1 vein. COMMENT

In reviewing our experience with these five cases certain features attracted our particuIar interest: First, the diagnosis of femoraI vein thrombosis when it is not associated with obvious thrombophIebitis-and the cases not associated with obvious thrombophIebitis are certainIy the most dangerous from the point of view of puImonary emboIism-is often diffrcuh to make because of the minor character of the clinica

EmboIism

AUGUST,194s

symptoms. If the possibiIity is kept in mind, however, the diagnosis can often be suspected and the suspicion verified by the use of venography. Positive treatment can then be instituted before a possibIe fatal puImonary emboIus occurs. In this connection it is worth noting that in one of our cases in which the diagnosis was suspected no further procedures were carried out because of the minor nature of the symptoms with the resuIt that this patient suffered an aImost fata emboIus at a later date (at which time the diagnosis of femora1 thrombosis was made and femora1 Iigation and resection successfuIIy performed). In cases in which a puImonary emboIus aIready has occurred and there is a good chance of another emboIus occurring, evidence of femora1 vein thrombosis should be carefuIIy Iooked for and even if no evidence is found venography shouId be done. In one of our patients there were absoIuteIy no cIinica1 signs suggesting any troubIe with the femoral vein or its tributaries but a thrombus was Iocated by venography and Iater removed at operation. SecondIy, the question of the relationship of edema of the Ieg to femora1 vein thrombosis interested us especially. In these few cases it was quite evident that marked edema was present only when there was definite thrombophIebitis in the periphera1 veins and was very sIight or absent when there was a bIand femora1 thrombosis, even though the femora1 vein was compIeteIy obstructed up to the saphenous junction. PostoperativeIy, there was sIight edema of the Iower Ieg for a few days in four of the five patients and in a11 of them this was associated with signs of some mild thrombophIebitis, as evidenced particuIarIy by some tenderness aIong the femoral vein. InterestingIy enough, in the one case in which marked thromboand periat operation the phIebitis was found edema disappeared entireIy a few days after femoral Iigation and partia1 resection. This probabIy was due to the interruption of the perivascuIar sympathetic nerve fibers produced by the resection of the femora1 vein. It is evident, therefore, that

NEW SERIES VOL.LXIX, No. 2

P&her-PuImonary

simple obstruction of the femoraI vein does not produce edema; and if the saphenous or other coIIatera1 systems are patent, femoral Iigation shouId not by itseIf cause edema. Furthermore we suspect that by resecting a small portion of the femoral vein at the time that we do a femoral ligation the probabiIity of postoperative edema of the Iower leg is greatly decreased in the same manner that thrombophIebitic edema is reheved. In fact, if Ochsner and DeBakey’s theory of thrombophIebitis’ is correct, interruption of the perifemora1 sympathetic chain should be good insurance against the postoperative development of a periphera1 thrombophIebitis, which as we have noted above is the chief cause of edema in these cases of venous thrombosis. However, our group of cases is too small to permit drawing any definite concIusion on this point. A third feature of interest to us in these cases, particuIarIy in the patients who had aIready suffered a puImonary emboIus, was the tremendous satisfaction we derived from being abIe to offer these patients rationa form of treatment a positive, instead of having to offer them and their families weeks of anxiety and danger with the possibiIity of sudden vioIent death threatening. The peace of mind obtained, in addition to the satisfying cIinica1 resuIts, was certainIy worth the sIight dangers offered by venography or the minor IocaI operation. Th e contrast between the oId and the new method of treatment is brought home forcibIy when a patient who has had a near fata puImonary emboIus is able to leave the hospita1 waIking two weeks Iater instead of spending anxious weeks in bed and not even being sure then that another emboIus may not occur. SUMMARY I. The etioIogy and present methods of treatment of puImonary emboli are reviewed, emphasizing the fact that most emboli are believed to originate from free propagating thrombi in the femoral venous system.

Embolism

American Journalofsurgery

203

2. The use of femora1 vein Iigation in the prophyIactic treatment of puImonary emboIism and the use of venography in the diagnosis and localization of femoral thrombi are discussed and evaIuated. in which 3. Five cases are reported accurate diagnosis of femoral vein thrombi was made IargeIy through the use of venography and in which satisfactory resuits were obtained by Iigating and partiaIly resecting the invoIved femoral vein. Three of the patients had suffered puImonary emboli before the femoral vein ligation was performed. 4. In addition to the remova of danger of further pulmonary emboIi, the operation of Iigation and partial resection of the femoraI vein was observed to have a beneficia1 effect on associated thrombophIebitis when this was present, particuIarIy in the prevention as we11 as relief of associated peripheral edema.

REFERENCES

N. W., NYGAARD, K. K., WALTERS, WALTMAN and PRIESTLEY, J. T. A statistica study of postoperative venous thrombus and puImonary embolism. Proc. Slafl Meet., Mayo Clin., 15: 796, 1940; 16: 33, 1941. 2. BAU~R, G. A venographic study of thromboembolic problems. Acta chir. Scandin&., 84: I, 1940. 3. BARKER, N. W., ALLEN, E. V. and WAUGH, J. M. The use of dicumarol in the prevention of postoperative thrombophIebitis and puImonary embolism. Proc. .%a$Meet., Mayo (%?I., 18: 102: 1943. 4. DOCGHERTY, JOHN and HOMANS,JOHN: Venography: a clinical study. Surg., Gynec. &+ Obst., 71: 697, I. BARKER,

1940. 5. STARR, ARNOLD, FRANK, H. A. and FINE, JACOB.

The venographic diagnosis of thrombophIebitis of the lower extremities. J. A. M. A., I 18: ,192, 1942. 6. ALLEN, ARTHUR W., LINTON, ROBERT R. and DONALDSON, GORDON A. Thrombosis and emboIism-review of zoz patients treated by femoral vein interruption. Ann. Surg., I 18: 728, 1943. 7. OCHSNER, ALTON and DEBAKEY, M. E. Therapeutic considerations of thrombophlebitis and phIebothrombosis. New England J. Med., 225: 207, 1941. 8. DEBAKEY, ~IICHAEL E., SCHROEDER, GEORGE F. and OCHSNER, ALTON. Significance of phlebography in phlebothrombosis. J. A. M. A., 123: 73% 1944. SANTOS, REYNALDO. o. DOS

veine. Caveinferieure 587. 1935.

Phlebonraohie dune suturee. J. d’wh., 39: 586-