- Email: [email protected]
EFFECT OF INTERMITTENT COMPRESSION OF THE ARMS ON DEEP VENOUS THROMBOSIS IN THE LEGS
1265 used in this study and D-locus typing, except that 102 shows a significant association with DW4.22 BT The mechanism of the association between HLA and M.S. remains speculative and has been discussed elsewhere.23-26 Clearly, before any firm conclusions can be drawn it is necessary to identify as precisely as possible which of the genes in the HLA region is most highly assera
susceptibility and what are the physiological functions of the particular gene product. This study confirms that the presence of certain B-lymphocyte alloantigens is associated with a greater relative risk of susceptibility to M.S. than that reported for any of the well-defined HLA antigens, including the DW2 deter-
EFFECT OF INTERMITTENT COMPRESSION OF THE ARMS ON DEEP VENOUS THROMBOSIS IN THE LEGS M. T. N. KNIGHT
Department of Surgery, Royal Postgraduate Medical School, Hammersmith Hospital, London W12
sociated with
determine whether M.S. B-lymphocyte alloantigens show the same associations in different ethnic groups in order to decide whether the B-lymphocyte alloantigens have a direct role in pathogenesis of M.S. or are merely the products of genes in linkage disequilibrium with the "M.S. suscep-
minant. It will be necessary
to
and
tibility gene".
assistance.
Requests for reprints should be
sent
J.R.B., Queen Victoria Hospi-
tal, East Grinstead, Sussex, RH19 3DZ.
REFERENCES 1. Svejgaard, A.,
Hauge, M., Jersild, C., Platz, P., Ryder, L. P., Staub Nielson, L., Thomsen, M. in Monographs in Human Genetics; vol. VII, The HLA System. Basle, 1975. 2. Bodmer, W. F. in Cytogenetics and Cell Genetics. New York, 1976. 3. Sachs, D. H., Cone, J. L. J. exp. Med. 1973, 138, 1289. 4. Schreffler, D. C., David, C. S. in Advances in Immunology (edited by F. J. D. Dixon and H. G. Kunkel); vol. xx. New York, 1975. 5. Benacerref, B., McDevitt, H. O. Science, 1972, 175, 273. 6. Jersild, C., Svejgaard, A., Fog, T. Lancet, 1972, i, 1242. 7. Naito, S., Namerou, N., Mickey, M. R., Terasaki, P. I. Tissue Antigens, 1972, 2, 1. 8. Jersild, C., Dupont, B., Fog, T., Platz, P. J., Svejgaard, A. Transplant. Rev. 148.
9.
Jersild, C., Hansen, G. S., Svejgaard, A., Fog, T., Thomsen, M., Dupont, B. Lancet, 1973, ii, 1221. 10. Bodmer, W. F. in Histocompatibility Testing (edited by J. Dausset and J. Colombani); p. 611. Copenhagen, 1972. 11. Winchester, R. J., Ebess, G., Fu, S. M., Espinosa, L., Zabriskie, J., Kunkel, H. G. Lancet, 1975, ii, 814. 12. Wernet, P. Transplant. Rev. 1976, 30, 294. 13. McDonald, W. I. in Multiple Sclerosis Research (edited by A. N. Davison, J. H. Humphrey, A. L. Liversedge, W. I. McDonald, and J. S. Porterfield); p. 1. H.M. Stationery Office, London, 1975. 14. McDonald, W. I., Halliday, A. M. Br. med. Bull. (in the press). 15. Batchelor, J. R., in Handbook of Experimental Immunology (edited by D. M. Weir); vol. II. Oxford, 1973. 16. Böyum, A. Scand. J. clin. Lab. Invest. 1968, 21, suppl., 97. 17. Wilson, A. B., Haegert, D. G., Coombs, R. R. A. Clin. exp. Immun. 1975, 22, 177. 18. Welsh, K. I., Turner, M. J. Tissue Antigens 1976, 8, 197. 19. Carp, R. I., Licursi, P. C., Merz, P. A., Merz, G. S. J. exp. Med. 1972, 136, 618. 20. Koldovsky, U., Koldovsky, P., Henle, W., Ackerman, R., Haase, G. Infect. Immun. 1975,12, 1355. 21. Henle, G., Koldovsky, U., Koldovsky, P., Henle, W., Ackerman, R., Haase, G. ibid. p. 1367. 22. Bodmer, J., Pickbourne, P., Bodmer, W., et al. Tissue Antigens (in the
press). 23. Bodmer, W. F. Nature, 1972, 237, 139. 24. Lancet, 1975, ii, 536. 25. McDevitt, H. O., Bodmer, W. F. ibid. 1974, i, 1269. 26. Batchelor, J. R. Br. med. Bull. (in the press).
Addendum Since this article was submitted, a study has been published (Terasaki, P. I., Park, M. S., Opelz, G., Ting, A. Science, 1976,193, 1245) which also describes a significantly increased frequency of a B-lymphocyte alloantigen in 56 patients with
unclassified M.s.
Despite the presence of venostasis in legs, intermittent compression of
the the arms during and after surgery reduced the incidence of deep venous thrombosis (D.V.T.) in the legs to half that in control patients and maintained blood fibrinolytic activity at preoperative values. It is suggested that the release of fibrinolytic activators is essential to the prophylactic action of pneumatic leggings.
Summary
Introduction AN early clinical trial by Hills et al.’ showed that the of intermittent-compression leggings (Flowtron-Aire Limited) reduced the incidence of postoperative D.v.T. from 40% to 5% in cancer-free patients, but did not reduce the incidence in cancer patients. This anomaly suggested that some mechanism other than simple pre7 vention of stasis was involved in the prophylactic action of the leggings. Allenby et awl. examined the blood-flow in the hind leg of a greyhound and found that the leggings, while emptying the great veins intermittently, also caused venous occlusion. Venous occlusion causes the release of fibrinolytic activators into the bloodand direct compression of a vein wall releases fibrinolytic activity into free flowing blood.4 Allenby examined the blood of patients wearing the leggings and found that although fibrinolytic activity is normally inhibited after operation, in cancer-free patients wearing the leggings fibrinolytic activity progressively increased. Despite application of the leggings, fibrinolytic activity declined after operation in patients with cancer. To determine whether these changes in fibrinolytic activity were essential to the protective function of the leggings or were merely an incidental finding the source of fibrinolytic activity was separated from the area of venous stasis and clot formation. Application of compression to the arms was chosen because fibrinolytic activity is known to be high in this area. 56 The aim was to establish whether intermittent compression of the arms enhanced blood fibrinolytic activity after operation and whether this would alter the incidence of D.V.T. in the legs. In this way any change in the incidence of D.V.T. would reflect the action of fibrinolytic substances in the blood and the importance of fibrinolytic activity to the protective action of the leggings would be confirmed or refuted. use
We thank the Medical Research Council and the East Grinstead Research Trust for financial support for this work; we also thank Mrs J. Hirsch, Mrs C. Seymour, and Mrs L. Simmons for expert technical
1975, 22,
ROBIN DAWSON
Patients and Methods 128 cancer-free patients over the age of fifty, subject to specific exclusions were invited to take part in the trial. Patients who had a history of D.v.T., pulmonary embolism, thyroid or renal disease, varicose veins, diabetes mellitus, and iodine allergy were excluded, as were premenopausal women. Informed written consent was recorded in a numbered trial register. Correspondingly numbered envelopes were opened to reveal the random allocation of patients to compression or control groups. 7 patients were withdrawn for the following reasons : 3 were discharged before leg scans were completed. 1
1266
patient in
the control group died during operation; 2 patients found to have carcinoma at operation; and 1 patient was given subcutaneous heparin as premedication.
COMPARISON OF GROUP CHARACTERISTICS
were
Armlets
Inflatable fabric sleeves extending from the wrist to the axilla were applied to both arms. A pump (Flowtron-Aire Limited), alternately inflated each armlet for 110 seconds to a plateau pressure of 50 mm Hg, followed by 130 seconds of deflation. Intermittent compression was started when premedication was given, was continued during operation, and for 24 hours postoperatively.
Venepuncture Antecubital venepuncture with a 19G 50 mm needle was carried out between 8 A.M. and 9 A.M. in patients who for twenty minutes beforehand had avoided smoking and taking caffeine or exercise. 20 mm Hg of sphygmomanometer cuff pressure was used to aid entry into the veins. A delay of twenty minutes was allowed between cessation of intermittent compression and blood sampling in the treated group on the first postoperative morning, to allow the removal of occluded blood from the arm and the restoration of uninterrupted blood-flow. 4.5ml of blood was added to 0.55 ml of 3-8% trisodium citrate and 3% HEPES buffer set in ice and water. Blood-samples were taken preoperatively and repeated on the first three postoperative mornings.
Euglobulin
Clot Lysis-time
Within twenty minutes of sampling the specimen was centrifuged at 1200 g at 4°C for twenty minutes. Duplicate solutions were made by adding 1 ml of platelet-free plasma to 9 ml of distilled water on ice to which 0.32 ml of 1% acetic acid had been added. After mixing, the solution was allowed to stand for ten to fifteen minutes before being centrifuged at 1200 g for three minutes. The supernatant was discarded, the centrifuge tube dried, and the deposit resuspended in 1 ml of imidazole buffer (pH 7-4). The solution was transferred to 10 cm x 5 mm glass tube and clotted with 5 N.I.H. units of thrombin in 0.ml of imidazole buffer and then placed in a waterbath at 37°C. Clot lysis was observed during the first two hours and then recorded photographically at 15-minute intervals. The time in minutes taken for the clots to disappear completely was recorded as the lysis-time. The results were expressed in more appropriate units of fibrinolytic activity-i.e., as the reciprocal of the square of the clot lysis-time multiplied by 106 for ease of expression.7
Plasma-fibrinogen Fibrinogen was assayed according mylen et a1.8 with a coagulometer.
to
the groups could be made on the type or duration of operations included, nor in the proportion of smokers involved (see table). There were however 16 females in the treated group and only 13 females in the control group, but all these patients were postmenopausal and their fibrinolytic activity was therefore similar to that of men of similar age.
"Resting" and Postoperative Fibrinolytic Activity Fibrinolytic activity is inversely proportional to the euglobulin clot lysis-time, but after conversion of the lysis-times to more appropriate units by taking the reciprocal of the square of the lysis-time x 106 low levels of fibrinolytic activity are represented by small values in units. In the control group there was a pronounced decrease in mean fibrinolytic activity immediately after operation with some slight return to normal values by the third postoperative day (fig. 1). The fibrinolytic activity in the treated group was significantly higher than that of the control group until the third postoperative day, the armlets significantly enhancing postoperative fibrinolytic
the method of Ver-
125[ Leg Scans 180 mg of potassium iodide was prescribed daily preoperatively and postoperatively for three weeks. Within an hour of operation 100[ Ci of 1-labelled fibrinogen was given intravenously and twenty minutes later baseline counts were taken of the heart and points marked out at 4-inch intervals from the ankle of each leg. A J. & P. Engineering scintillation counter was used and duplicate background readings were taken. Patients were scanned while recumbent, with legs raised to 300 from the horizontal every twenty-four hours for 5 postoperative days. The criteria of Browse and Negus9 were used to interpret the computed percentage uptakes.
Results
Comparison of Group Characteristics The groups
weight (see
well-matched for height, age, and accompanying table). No distinction between were
Fig. 1--Changes groups.
in
fibrinolytic activity of control
and treated
1267 the activity remained significantly above that of the control group for at least eighteen hours after intermittent
compression was stopped. known how the fibrinolytic activity in repeatedly compressed arms is related to that found in other areas of systemic circulation. Although our findings are not a direct measure of systemic levels of fibrinolytic activity they do indicate a significant and persistent increase in the fibrinolytic activity of free flowing blood in the arms subjected to intermittent compression. The precise effect of intermittent compression on systemic fibrinolytic activity is the subject of a further study. It has been stated that the postoperative lengthening of the euglobulin clot lysis-time merely reflects the increased fibrinogen content of the plasma euglobulin fraction after operation." Despite the considerable increase in the fibrinogen levels there is no significant difference between the mean fibrinogen levels of the two groups before or after operation. Therefore the highly significant differences in the euglobulin clot lysis-times of the two groups reflect real differences in fibrinolytic activity and not a difference in plasma fibrinogen conIt is
not
tent.
The frequency of D.v.T. in the legs was reduced by than half by the application of intermittent compression to the arms. This is evidence of an action of the device which is independent of local mechanical effects and which is effective despite the presumed presence of venous stasis and lack of venous pulsatility12 in the legs. more
Fig. 2-Changes
in
plasma-fibrinogen
in control and treated
groups.
activity by maintaining the mean activity at the preoperative level. This enhancement persisted for at least eighhours after removal of the armlets. the effect of differences in plasma in two groups, concurrent plasma the fibrinogen concentrations were measured. Despite the fibrinogen considerable increase in fibrinogen concentration after operation there was no significant difference in plasmafibrinogen in the two groups before or after operation (fig. 2). Comparison of figs. 1 and 2 indicates that changes in the euglobulin clot lysis-time reflect a real change in fibrinolytic activity and not a difference in the fibrinogen content of the plasma. D.v.T. developed in 31.6% (19/60) of the control group and 13-6% (8/61) of the treated group (P=0.029). 26 (21.6%) of the legs of control patients and 12 (9.8%) of the legs of those receiving intermittent compression of the arms were affected (P=0020). The significance of these results was calculated by the x.2 test with Yates’ correction. Application of the armlets reduced the incidence of leg D.V.T. by more than half. teen
To exclude
Discussion The reduction in fibrinolytic activity which normally follows surgical operation was prevented by intermittent compression of the arms. In the control group mean fibrinolytic activity was reduced immediately after operation whilst in patients wearing the armlets this activity did not fall until the third postoperative day. The difference in activity between the two groups during the first
forty-eight
hours after operation was highly significant. Intermittent compression of the arms maintained mean fibrinolytic activity at or above the preoperative value during the period of intermittent compression and
The pronounced increase in fibrinolytic.activity during the early stages of an operation followed by the decline from peak activity before the end of the operation 13 and the prolonged diminution of activity after operationI4-I7 suggest that either stores of fibrinolytic activity have been depleted or that perhaps the mechanism of release has changed during the period of operation. A possible explanation for the action of the armlets is that repeated occlusion and massage of the arm veins causes the local release of plasminogen activator which in turn overflows into the systemic circulation. Sufficient plasminogen activator and plasmin eventually escape inhibition and removal from the circulation to reach the legs. Here they make good the post-traumatic deficit in production noted by Rawles et al. 18 and restore the necessary balance between coagulation and fibrinolysis first postulated by Fearnleyl9 and Astrup.2° Once the rate of thrombosis formation is matched by its rate of clearance thrombosis is unlikely to become established. The release of plasminogen activators and the reduction of distant thrombosis despite the presence of venous stasis and systemic dilution suggest that the release of fibrinolytic activity is essential to the prophylactic action of pneumatic leggings. We thank Prof. D. G. Melrose for advice and encouragement the study, Prof. J. S. Calnan for his help; Prof. R. B. Welbourn, Mr J. Spencer, and Mr J. L. l’oak for allowing us to study patients in the surgical wards; and Miss L. Jones and Miss R. Ackerley for early technical assistance. This study was supported by a grant from the Department of Health and Social Security.
throughout
Requests for reprints should be addressed to M.T.N.K. REFERENCES 1
Hills, N H., Pflug, J. J., Jeyasingh, K., Boardman, L., Calnan, J. S. Br. med. J. 1972, i, 131. 2. Allenby, F. Boardman, L., Pflug, J. J., Calnan, J. S. Lancet, 1973, ii, 1412 3 Clarke, R. L., Orandi, A., Clifton, E. E. Angiology, 1960,11, 367.
1268
TREATMENT OF HYPERSPLENISM BY EMBOLUS PLACEMENT IN THE SPLENIC ARTERY C. TRITAKIS
J. PAPADIMITRIOU G. KARATZAS
2nd Surgical Department and Department of Roentgenology, Athens University, Aretaieion Hospital, Athens
A. PAPAIOANNOU 2nd Surgical Unit,
Evangelismos Hospital, Athens
patient with liver cirrhosis and hypersplenism resistant to corticosteroids splenectomy was attempted but proved impossible. Embolisation of the splenic artery with tiny fragments of absorbable gelatin sponge induced gradual restitution of hæmatological function to normal after 6 wk. Abdominal pain, paralytic ileus of short duration, transient pyrexia, and pleural effusion ensued but were well tolerated. The patient has remained well, 2 1/2 mo later. It is suggested that this simple method may prove rewarding and could safely be used, if necessary, more than once in patients for whom other treatments are un-
Summary
In
a
severe
have moderate ascites and splenomegaly. Hasmatocrit was 37, white blood-cells 320/mm3, platelets rare. The bonemarrow was very cellular with numerous megakaryocytes and increased responses from the white-cell and red-cell series. The b_romsulphthalein test was slightly elevated, and the bloodammonia was 315 mg/dl. Colloidal-gold liver and spleen scanning confirmed splenomegaly. The albumin/globulin ratio, alkaline phosphatase, serum-bilirubin, and transaminases were within normal limits. A course of prednisone failed to reduce hypersplenism, and splenectomy was decided upon after intensive correction of anxmia and thrombocytopenia with fresh blood and platelet transfusions. An exploration was attempted on June 9, 1976, but was quickly abandoned because of massive blood-loss due to extensive portosystemic shunts and
bleeding tendency. The patient was admitted to Aretaieion Hospital and was catheterised through the right femoral artery on July 27. Selective angiography of the coeliac artery was done, and the’ hepatic artery was not seen (fig. 1). In another angiogram the hepatic artery was seen to branch off the superior mesenteric
suitable or have failed.
Introduction SPLENECTOMY is the usual treatment of choice for in patients who are too ill or for other reasons are unable to undergo this procedure, different therapeutic measures must be used. Experimental evidence suggests that introduction of emboli into the splenic artery is without serious adverse effectsI and that complete dearterialisation of the spleen can suppress splenic function without causing necrosis.2 We report successful embolisation of the splenic artery in a patient with severe hypersplenism in whom splenectomy
hypersplenism. However,
was
impossible. Case-report
A
52-year-old woman presented with a 2 mo history of progressive ascites and weakness, splenomegaly with striking hypersplenism, and, more recently, haemoptysis. She had had jaundice at age 35, three abdominal operations for intestinal obstruction, and, in the past 4 yr, two episodes of allegedly massive intestinal bleeding from an unidentified source. On admission to Evangelismos Hospital, she was found to
Fig. I-Angiogram of
cfeliac artery.
Hepatic artery not seen, splenic artery very wide and tortuous. Numbers refer to main branches of splenic artery displayed on arteriography and subsequently obstructed.
4. 5.
Chakrabarti, R., Binrs, P. M., Fearnley, G. R. Lancet, 1963, i, 1289. Pandolfi, M., Robertson, B., Isacson, S., Nilsson, I. M. Thromb. Diath. hœmorrh. 1968, 20, 247. 6. Robertson, B. R., Pandolfi, M., Nilsson, I. M. Acta chir. scand. 1972, 138, 437.
Merrils, R. J., Shaw, J. T. R. Biochem. J. 1968, 8, 418. Vermylen, C., De Vrcker, R. A., Verstraete, M. Clinica chim. Acta, 1963, 8, 418 9. Browse, N. L., Negus, D. Br. med. J. 1970, iii, 615. 10. Gordon-Smith, I. C., Hickman, J. A., Le Quesne, L. P. Br. J. Surg. 1974, 61, 213. 11. Hickman, J. A., Gordon-Smith, I. C., Whitfield, P. F., Godfrey, J. S. J. clin. Path. 1973, 26, 189. 12. Roberts, V. C., Sabri, S., Beeley, A. H., Cotton, L. T. Br. J. Surg. 1972, 59, 223. 13. Mansfield, A. O. Br. J. Surg. 1972, 59, 754. 14. Chakrabarti, R., Hockin, E. D., Fearnley, G. R. J. clin. Path. 1969, 22, 650. 15. Saulter, R. D., Myers, W. O., Jefferson, F. R. III, Wenzel, F. J. Archs. Surg. 1973, 107, 292. 16. Bennett, N. B., Ogston, C. M., Ogston, D. Clin. Sci. 1967, 32, 27. 17. Ygge, J. Am. J. Surg. 1970, 119, 225. 18 Rawles, J. M., Warlow, C., Ogston, D. Br. med. J. 1975, ii, 61. 19. Fearnley, G. R. ibid. 1961, i, 992. 20. Astrup, T. Lancet, 1956, ii, 565. 7. 8.
-
11
Fig. 2-Angiogram ofsuperior mesenteric artery, demonstrating origin of hepatic artery.
susceptibility and what are the physiological functions of the particular gene product. This study confirms that the presence of certain B-lymphocyte alloantigens is associated with a greater relative risk of susceptibility to M.S. than that reported for any of the well-defined HLA antigens, including the DW2 deter-
EFFECT OF INTERMITTENT COMPRESSION OF THE ARMS ON DEEP VENOUS THROMBOSIS IN THE LEGS M. T. N. KNIGHT
Department of Surgery, Royal Postgraduate Medical School, Hammersmith Hospital, London W12
sociated with
determine whether M.S. B-lymphocyte alloantigens show the same associations in different ethnic groups in order to decide whether the B-lymphocyte alloantigens have a direct role in pathogenesis of M.S. or are merely the products of genes in linkage disequilibrium with the "M.S. suscep-
minant. It will be necessary
to
and
tibility gene".
assistance.
Requests for reprints should be
sent
J.R.B., Queen Victoria Hospi-
tal, East Grinstead, Sussex, RH19 3DZ.
REFERENCES 1. Svejgaard, A.,
Hauge, M., Jersild, C., Platz, P., Ryder, L. P., Staub Nielson, L., Thomsen, M. in Monographs in Human Genetics; vol. VII, The HLA System. Basle, 1975. 2. Bodmer, W. F. in Cytogenetics and Cell Genetics. New York, 1976. 3. Sachs, D. H., Cone, J. L. J. exp. Med. 1973, 138, 1289. 4. Schreffler, D. C., David, C. S. in Advances in Immunology (edited by F. J. D. Dixon and H. G. Kunkel); vol. xx. New York, 1975. 5. Benacerref, B., McDevitt, H. O. Science, 1972, 175, 273. 6. Jersild, C., Svejgaard, A., Fog, T. Lancet, 1972, i, 1242. 7. Naito, S., Namerou, N., Mickey, M. R., Terasaki, P. I. Tissue Antigens, 1972, 2, 1. 8. Jersild, C., Dupont, B., Fog, T., Platz, P. J., Svejgaard, A. Transplant. Rev. 148.
9.
Jersild, C., Hansen, G. S., Svejgaard, A., Fog, T., Thomsen, M., Dupont, B. Lancet, 1973, ii, 1221. 10. Bodmer, W. F. in Histocompatibility Testing (edited by J. Dausset and J. Colombani); p. 611. Copenhagen, 1972. 11. Winchester, R. J., Ebess, G., Fu, S. M., Espinosa, L., Zabriskie, J., Kunkel, H. G. Lancet, 1975, ii, 814. 12. Wernet, P. Transplant. Rev. 1976, 30, 294. 13. McDonald, W. I. in Multiple Sclerosis Research (edited by A. N. Davison, J. H. Humphrey, A. L. Liversedge, W. I. McDonald, and J. S. Porterfield); p. 1. H.M. Stationery Office, London, 1975. 14. McDonald, W. I., Halliday, A. M. Br. med. Bull. (in the press). 15. Batchelor, J. R., in Handbook of Experimental Immunology (edited by D. M. Weir); vol. II. Oxford, 1973. 16. Böyum, A. Scand. J. clin. Lab. Invest. 1968, 21, suppl., 97. 17. Wilson, A. B., Haegert, D. G., Coombs, R. R. A. Clin. exp. Immun. 1975, 22, 177. 18. Welsh, K. I., Turner, M. J. Tissue Antigens 1976, 8, 197. 19. Carp, R. I., Licursi, P. C., Merz, P. A., Merz, G. S. J. exp. Med. 1972, 136, 618. 20. Koldovsky, U., Koldovsky, P., Henle, W., Ackerman, R., Haase, G. Infect. Immun. 1975,12, 1355. 21. Henle, G., Koldovsky, U., Koldovsky, P., Henle, W., Ackerman, R., Haase, G. ibid. p. 1367. 22. Bodmer, J., Pickbourne, P., Bodmer, W., et al. Tissue Antigens (in the
press). 23. Bodmer, W. F. Nature, 1972, 237, 139. 24. Lancet, 1975, ii, 536. 25. McDevitt, H. O., Bodmer, W. F. ibid. 1974, i, 1269. 26. Batchelor, J. R. Br. med. Bull. (in the press).
Addendum Since this article was submitted, a study has been published (Terasaki, P. I., Park, M. S., Opelz, G., Ting, A. Science, 1976,193, 1245) which also describes a significantly increased frequency of a B-lymphocyte alloantigen in 56 patients with
unclassified M.s.
Despite the presence of venostasis in legs, intermittent compression of
the the arms during and after surgery reduced the incidence of deep venous thrombosis (D.V.T.) in the legs to half that in control patients and maintained blood fibrinolytic activity at preoperative values. It is suggested that the release of fibrinolytic activators is essential to the prophylactic action of pneumatic leggings.
Summary
Introduction AN early clinical trial by Hills et al.’ showed that the of intermittent-compression leggings (Flowtron-Aire Limited) reduced the incidence of postoperative D.v.T. from 40% to 5% in cancer-free patients, but did not reduce the incidence in cancer patients. This anomaly suggested that some mechanism other than simple pre7 vention of stasis was involved in the prophylactic action of the leggings. Allenby et awl. examined the blood-flow in the hind leg of a greyhound and found that the leggings, while emptying the great veins intermittently, also caused venous occlusion. Venous occlusion causes the release of fibrinolytic activators into the bloodand direct compression of a vein wall releases fibrinolytic activity into free flowing blood.4 Allenby examined the blood of patients wearing the leggings and found that although fibrinolytic activity is normally inhibited after operation, in cancer-free patients wearing the leggings fibrinolytic activity progressively increased. Despite application of the leggings, fibrinolytic activity declined after operation in patients with cancer. To determine whether these changes in fibrinolytic activity were essential to the protective function of the leggings or were merely an incidental finding the source of fibrinolytic activity was separated from the area of venous stasis and clot formation. Application of compression to the arms was chosen because fibrinolytic activity is known to be high in this area. 56 The aim was to establish whether intermittent compression of the arms enhanced blood fibrinolytic activity after operation and whether this would alter the incidence of D.V.T. in the legs. In this way any change in the incidence of D.V.T. would reflect the action of fibrinolytic substances in the blood and the importance of fibrinolytic activity to the protective action of the leggings would be confirmed or refuted. use
We thank the Medical Research Council and the East Grinstead Research Trust for financial support for this work; we also thank Mrs J. Hirsch, Mrs C. Seymour, and Mrs L. Simmons for expert technical
1975, 22,
ROBIN DAWSON
Patients and Methods 128 cancer-free patients over the age of fifty, subject to specific exclusions were invited to take part in the trial. Patients who had a history of D.v.T., pulmonary embolism, thyroid or renal disease, varicose veins, diabetes mellitus, and iodine allergy were excluded, as were premenopausal women. Informed written consent was recorded in a numbered trial register. Correspondingly numbered envelopes were opened to reveal the random allocation of patients to compression or control groups. 7 patients were withdrawn for the following reasons : 3 were discharged before leg scans were completed. 1
1266
patient in
the control group died during operation; 2 patients found to have carcinoma at operation; and 1 patient was given subcutaneous heparin as premedication.
COMPARISON OF GROUP CHARACTERISTICS
were
Armlets
Inflatable fabric sleeves extending from the wrist to the axilla were applied to both arms. A pump (Flowtron-Aire Limited), alternately inflated each armlet for 110 seconds to a plateau pressure of 50 mm Hg, followed by 130 seconds of deflation. Intermittent compression was started when premedication was given, was continued during operation, and for 24 hours postoperatively.
Venepuncture Antecubital venepuncture with a 19G 50 mm needle was carried out between 8 A.M. and 9 A.M. in patients who for twenty minutes beforehand had avoided smoking and taking caffeine or exercise. 20 mm Hg of sphygmomanometer cuff pressure was used to aid entry into the veins. A delay of twenty minutes was allowed between cessation of intermittent compression and blood sampling in the treated group on the first postoperative morning, to allow the removal of occluded blood from the arm and the restoration of uninterrupted blood-flow. 4.5ml of blood was added to 0.55 ml of 3-8% trisodium citrate and 3% HEPES buffer set in ice and water. Blood-samples were taken preoperatively and repeated on the first three postoperative mornings.
Euglobulin
Clot Lysis-time
Within twenty minutes of sampling the specimen was centrifuged at 1200 g at 4°C for twenty minutes. Duplicate solutions were made by adding 1 ml of platelet-free plasma to 9 ml of distilled water on ice to which 0.32 ml of 1% acetic acid had been added. After mixing, the solution was allowed to stand for ten to fifteen minutes before being centrifuged at 1200 g for three minutes. The supernatant was discarded, the centrifuge tube dried, and the deposit resuspended in 1 ml of imidazole buffer (pH 7-4). The solution was transferred to 10 cm x 5 mm glass tube and clotted with 5 N.I.H. units of thrombin in 0.ml of imidazole buffer and then placed in a waterbath at 37°C. Clot lysis was observed during the first two hours and then recorded photographically at 15-minute intervals. The time in minutes taken for the clots to disappear completely was recorded as the lysis-time. The results were expressed in more appropriate units of fibrinolytic activity-i.e., as the reciprocal of the square of the clot lysis-time multiplied by 106 for ease of expression.7
Plasma-fibrinogen Fibrinogen was assayed according mylen et a1.8 with a coagulometer.
to
the groups could be made on the type or duration of operations included, nor in the proportion of smokers involved (see table). There were however 16 females in the treated group and only 13 females in the control group, but all these patients were postmenopausal and their fibrinolytic activity was therefore similar to that of men of similar age.
"Resting" and Postoperative Fibrinolytic Activity Fibrinolytic activity is inversely proportional to the euglobulin clot lysis-time, but after conversion of the lysis-times to more appropriate units by taking the reciprocal of the square of the lysis-time x 106 low levels of fibrinolytic activity are represented by small values in units. In the control group there was a pronounced decrease in mean fibrinolytic activity immediately after operation with some slight return to normal values by the third postoperative day (fig. 1). The fibrinolytic activity in the treated group was significantly higher than that of the control group until the third postoperative day, the armlets significantly enhancing postoperative fibrinolytic
the method of Ver-
125[ Leg Scans 180 mg of potassium iodide was prescribed daily preoperatively and postoperatively for three weeks. Within an hour of operation 100[ Ci of 1-labelled fibrinogen was given intravenously and twenty minutes later baseline counts were taken of the heart and points marked out at 4-inch intervals from the ankle of each leg. A J. & P. Engineering scintillation counter was used and duplicate background readings were taken. Patients were scanned while recumbent, with legs raised to 300 from the horizontal every twenty-four hours for 5 postoperative days. The criteria of Browse and Negus9 were used to interpret the computed percentage uptakes.
Results
Comparison of Group Characteristics The groups
weight (see
well-matched for height, age, and accompanying table). No distinction between were
Fig. 1--Changes groups.
in
fibrinolytic activity of control
and treated
1267 the activity remained significantly above that of the control group for at least eighteen hours after intermittent
compression was stopped. known how the fibrinolytic activity in repeatedly compressed arms is related to that found in other areas of systemic circulation. Although our findings are not a direct measure of systemic levels of fibrinolytic activity they do indicate a significant and persistent increase in the fibrinolytic activity of free flowing blood in the arms subjected to intermittent compression. The precise effect of intermittent compression on systemic fibrinolytic activity is the subject of a further study. It has been stated that the postoperative lengthening of the euglobulin clot lysis-time merely reflects the increased fibrinogen content of the plasma euglobulin fraction after operation." Despite the considerable increase in the fibrinogen levels there is no significant difference between the mean fibrinogen levels of the two groups before or after operation. Therefore the highly significant differences in the euglobulin clot lysis-times of the two groups reflect real differences in fibrinolytic activity and not a difference in plasma fibrinogen conIt is
not
tent.
The frequency of D.v.T. in the legs was reduced by than half by the application of intermittent compression to the arms. This is evidence of an action of the device which is independent of local mechanical effects and which is effective despite the presumed presence of venous stasis and lack of venous pulsatility12 in the legs. more
Fig. 2-Changes
in
plasma-fibrinogen
in control and treated
groups.
activity by maintaining the mean activity at the preoperative level. This enhancement persisted for at least eighhours after removal of the armlets. the effect of differences in plasma in two groups, concurrent plasma the fibrinogen concentrations were measured. Despite the fibrinogen considerable increase in fibrinogen concentration after operation there was no significant difference in plasmafibrinogen in the two groups before or after operation (fig. 2). Comparison of figs. 1 and 2 indicates that changes in the euglobulin clot lysis-time reflect a real change in fibrinolytic activity and not a difference in the fibrinogen content of the plasma. D.v.T. developed in 31.6% (19/60) of the control group and 13-6% (8/61) of the treated group (P=0.029). 26 (21.6%) of the legs of control patients and 12 (9.8%) of the legs of those receiving intermittent compression of the arms were affected (P=0020). The significance of these results was calculated by the x.2 test with Yates’ correction. Application of the armlets reduced the incidence of leg D.V.T. by more than half. teen
To exclude
Discussion The reduction in fibrinolytic activity which normally follows surgical operation was prevented by intermittent compression of the arms. In the control group mean fibrinolytic activity was reduced immediately after operation whilst in patients wearing the armlets this activity did not fall until the third postoperative day. The difference in activity between the two groups during the first
forty-eight
hours after operation was highly significant. Intermittent compression of the arms maintained mean fibrinolytic activity at or above the preoperative value during the period of intermittent compression and
The pronounced increase in fibrinolytic.activity during the early stages of an operation followed by the decline from peak activity before the end of the operation 13 and the prolonged diminution of activity after operationI4-I7 suggest that either stores of fibrinolytic activity have been depleted or that perhaps the mechanism of release has changed during the period of operation. A possible explanation for the action of the armlets is that repeated occlusion and massage of the arm veins causes the local release of plasminogen activator which in turn overflows into the systemic circulation. Sufficient plasminogen activator and plasmin eventually escape inhibition and removal from the circulation to reach the legs. Here they make good the post-traumatic deficit in production noted by Rawles et al. 18 and restore the necessary balance between coagulation and fibrinolysis first postulated by Fearnleyl9 and Astrup.2° Once the rate of thrombosis formation is matched by its rate of clearance thrombosis is unlikely to become established. The release of plasminogen activators and the reduction of distant thrombosis despite the presence of venous stasis and systemic dilution suggest that the release of fibrinolytic activity is essential to the prophylactic action of pneumatic leggings. We thank Prof. D. G. Melrose for advice and encouragement the study, Prof. J. S. Calnan for his help; Prof. R. B. Welbourn, Mr J. Spencer, and Mr J. L. l’oak for allowing us to study patients in the surgical wards; and Miss L. Jones and Miss R. Ackerley for early technical assistance. This study was supported by a grant from the Department of Health and Social Security.
throughout
Requests for reprints should be addressed to M.T.N.K. REFERENCES 1
Hills, N H., Pflug, J. J., Jeyasingh, K., Boardman, L., Calnan, J. S. Br. med. J. 1972, i, 131. 2. Allenby, F. Boardman, L., Pflug, J. J., Calnan, J. S. Lancet, 1973, ii, 1412 3 Clarke, R. L., Orandi, A., Clifton, E. E. Angiology, 1960,11, 367.
1268
TREATMENT OF HYPERSPLENISM BY EMBOLUS PLACEMENT IN THE SPLENIC ARTERY C. TRITAKIS
J. PAPADIMITRIOU G. KARATZAS
2nd Surgical Department and Department of Roentgenology, Athens University, Aretaieion Hospital, Athens
A. PAPAIOANNOU 2nd Surgical Unit,
Evangelismos Hospital, Athens
patient with liver cirrhosis and hypersplenism resistant to corticosteroids splenectomy was attempted but proved impossible. Embolisation of the splenic artery with tiny fragments of absorbable gelatin sponge induced gradual restitution of hæmatological function to normal after 6 wk. Abdominal pain, paralytic ileus of short duration, transient pyrexia, and pleural effusion ensued but were well tolerated. The patient has remained well, 2 1/2 mo later. It is suggested that this simple method may prove rewarding and could safely be used, if necessary, more than once in patients for whom other treatments are un-
Summary
In
a
severe
have moderate ascites and splenomegaly. Hasmatocrit was 37, white blood-cells 320/mm3, platelets rare. The bonemarrow was very cellular with numerous megakaryocytes and increased responses from the white-cell and red-cell series. The b_romsulphthalein test was slightly elevated, and the bloodammonia was 315 mg/dl. Colloidal-gold liver and spleen scanning confirmed splenomegaly. The albumin/globulin ratio, alkaline phosphatase, serum-bilirubin, and transaminases were within normal limits. A course of prednisone failed to reduce hypersplenism, and splenectomy was decided upon after intensive correction of anxmia and thrombocytopenia with fresh blood and platelet transfusions. An exploration was attempted on June 9, 1976, but was quickly abandoned because of massive blood-loss due to extensive portosystemic shunts and
bleeding tendency. The patient was admitted to Aretaieion Hospital and was catheterised through the right femoral artery on July 27. Selective angiography of the coeliac artery was done, and the’ hepatic artery was not seen (fig. 1). In another angiogram the hepatic artery was seen to branch off the superior mesenteric
suitable or have failed.
Introduction SPLENECTOMY is the usual treatment of choice for in patients who are too ill or for other reasons are unable to undergo this procedure, different therapeutic measures must be used. Experimental evidence suggests that introduction of emboli into the splenic artery is without serious adverse effectsI and that complete dearterialisation of the spleen can suppress splenic function without causing necrosis.2 We report successful embolisation of the splenic artery in a patient with severe hypersplenism in whom splenectomy
hypersplenism. However,
was
impossible. Case-report
A
52-year-old woman presented with a 2 mo history of progressive ascites and weakness, splenomegaly with striking hypersplenism, and, more recently, haemoptysis. She had had jaundice at age 35, three abdominal operations for intestinal obstruction, and, in the past 4 yr, two episodes of allegedly massive intestinal bleeding from an unidentified source. On admission to Evangelismos Hospital, she was found to
Fig. I-Angiogram of
cfeliac artery.
Hepatic artery not seen, splenic artery very wide and tortuous. Numbers refer to main branches of splenic artery displayed on arteriography and subsequently obstructed.
4. 5.
Chakrabarti, R., Binrs, P. M., Fearnley, G. R. Lancet, 1963, i, 1289. Pandolfi, M., Robertson, B., Isacson, S., Nilsson, I. M. Thromb. Diath. hœmorrh. 1968, 20, 247. 6. Robertson, B. R., Pandolfi, M., Nilsson, I. M. Acta chir. scand. 1972, 138, 437.
Merrils, R. J., Shaw, J. T. R. Biochem. J. 1968, 8, 418. Vermylen, C., De Vrcker, R. A., Verstraete, M. Clinica chim. Acta, 1963, 8, 418 9. Browse, N. L., Negus, D. Br. med. J. 1970, iii, 615. 10. Gordon-Smith, I. C., Hickman, J. A., Le Quesne, L. P. Br. J. Surg. 1974, 61, 213. 11. Hickman, J. A., Gordon-Smith, I. C., Whitfield, P. F., Godfrey, J. S. J. clin. Path. 1973, 26, 189. 12. Roberts, V. C., Sabri, S., Beeley, A. H., Cotton, L. T. Br. J. Surg. 1972, 59, 223. 13. Mansfield, A. O. Br. J. Surg. 1972, 59, 754. 14. Chakrabarti, R., Hockin, E. D., Fearnley, G. R. J. clin. Path. 1969, 22, 650. 15. Saulter, R. D., Myers, W. O., Jefferson, F. R. III, Wenzel, F. J. Archs. Surg. 1973, 107, 292. 16. Bennett, N. B., Ogston, C. M., Ogston, D. Clin. Sci. 1967, 32, 27. 17. Ygge, J. Am. J. Surg. 1970, 119, 225. 18 Rawles, J. M., Warlow, C., Ogston, D. Br. med. J. 1975, ii, 61. 19. Fearnley, G. R. ibid. 1961, i, 992. 20. Astrup, T. Lancet, 1956, ii, 565. 7. 8.
-
11
Fig. 2-Angiogram ofsuperior mesenteric artery, demonstrating origin of hepatic artery.