- Email: [email protected]
RANDOMISED TRIAL OF INTRAVENOUS RECOMBINANT TISSUE-TYPE PLASMINOGEN ACTIVATOR VERSUS INTRAVENOUS STREPTOKINASE IN ACUTE MYOCARDIAL INFARCTION
842
in a patient who had had 2 previous complications. The fact that the study was not double blind is also unlikely to have biased the results, although it may have influenced the frequency of caesarean sections. Unfortunately, we did not measure platelet aggregation in mothers, so we cannot be sure whether they complied with the drug regimen. Also, we did not check on whether women in the control group took aspirin during the pregnancy. The combination of dipyridamole and aspirin did not cause apparent harm to mother or fetus, although the safety of aspirin during pregnancy has been seriously questioned by workers who have reported a high frequency of abnormal bleeding at delivery and/or of mucosal or cutaneous haemorrhage in babies. 10 The absence of such complications in our series, although they were carefully looked for, suggests that the doses of aspirin we gave, which is lower than those usually given, are likely to be harmless to the pregnant mother and her baby.
RANDOMISED TRIAL OF INTRAVENOUS RECOMBINANT TISSUE-TYPE PLASMINOGEN ACTIVATOR VERSUS INTRAVENOUS STREPTOKINASE IN ACUTE MYOCARDIAL
INFARCTION
Report from the European Cooperative Study Group for Recombinant Tissue-type Plasminogen Activator* M. VERSTRAETE M. BORY D. COLLEN R. ERBEL
R. J. LENNANE D. MATHEY H. R. MICHELS M. SCHARTL R. UEBIS
Summary
R. BERNARD R. W. BROWER D. P. DE BONO W. HUHMANN J. LUBSEN J. MEYER W. RUTSCH W. SCHMIDT R. VON ESSEN
In a single-blind randomised trial in patients with acute myocardial infarction of less than
6 h duration, the frequency of coronary patency was found to be higher after intravenous administration of recombinant human tissue-type plasminogen activator (rt-PA) than after intravenous streptokinase. 64 patients were allocated to 0·75 mg rt-PA/kg over 90 min, and the infarct-related coronary artery was patent in 70% of 61 assessable coronary angiograms taken 75-90 min after the start of infusion; 65 patients were allocated to 1 500 000 IU streptokinase over 60 min, and the infarct-related vessel was patent in 55% of 62 assessable angiograms. The 95% confidence interval of the difference ranges from ±30 to -2% (p = 0·054). Bleeding *STEERING COMMITTEE: M. Verstraete (Leuven) (chairman); R. J. Lennane (Ingelheim) (clinical coordinator); D. P. de Bono (Edinburgh); J. Lubsen (Rotterdam); D. Mathey (Hamburg); R. von Essen (Aachen). ADVISORY BOARD: J. Hampton (Nottingham); H. J. Jesdinsky (Düsseldorf); D. G. Julian (Newcastle upon Tyne); W. Schaper (Bad Nauheim); L. Wilhelmsen (Göteborg). DATA CENTRE: J. Lubsen, R. W. Brower, P. Fioretti, B. Soward, M. Bokslag (Thoraxcentre, Rotterdam). ANGIOGRAPHY EVALUATION GROUP: D. P. de Bono (Edinburgh) (secretary); W. S. Hillis (Glasgow); D. Reid (Newcastle); C. Turnbull (Edinburgh). CENTRAL COAGULATION LABORATORY: D. Collen, H. R. Lijnen (Leuven). PARTICIPATING CLINICAL CENTRES: Innere Medizin I, RheinischWestfälische Technische Hochschule, Aachen (S. Effert, R. von Essen, R. Uebis, W. Schmidt); Academisch Medisch Centrum, Amsterdam (A. J. Dunning, K. Romyn); Klinikum Charlottenburg, Freie Universität Berlin (H.
Correspondence should be addressed to M. B., Service de Néphrologle, Hôpital Tenon, 4 Rue de la Chine, 75970 Paris, Cedex 20, France. REFERENCES
propos du poids et de la taille des nouveaux nés à la naissance. Rev Franc Gynec 1971; 66: 391-96. 2. Bonnar J, Redman CWG, Denson KW. The role of coagulation and fibrinolysis in preeclampsia. In- Lindheimer MD, Katz AI, Zuspan F, eds. Hypertension in pregnancy. New York: Wiley, 1976: 85-93. 3. Redman CWG, Bonner J, Beilin L. Early platelet consumption in pre-eclampsia Lancet 1978; i: 467-69. 4. McKay DG. Chronic intravascular coagulation in normal pregnancy and preeclampsia. Contributions Nephrol 1981; 25: 108-19. 5. Howie PW, Prentice CRM, Forbes CD. Failure of heparin therapy to affect the clinical course of severe preeclampsia. Br J Obstet Gynaecol 1979; 82: 711-17 6. Marcus AJ. Editorial retrospective: Aspirin as an antithrombotic medication N Engl J Med 1983; 309: 1515-17. 7 Salzman EW. Aspirin to prevent arterial thrombosis. N Engl J Med 1982; 307: 113-15. 8. Lewis HD, Davis JW, Archibald DG, et al. Protective effects of aspirin against acute myocardial infarction and death in men with unstable angina. Results of a Veterans Administration cooperative study. N Engl J Med 1983; 309: 396-403. 9. Persantin-Aspirin Reinfarction Study Research Group. Persantin and aspirin in coronary heart disease. Circulation 1980; 62: 449-61. 10 Stuart MJ, Gross SJ, Elrad H, Graeber JE. Effects of acetylsalicylic acid ingestion on maternal and neonatal hemostasis. N Engl J Med 1982; 307: 909-12. 1.
Leroy B, Lefort F. A
episodes and other complications were less common in the rt-PA patients than in the streptokinase group. Hospital mortality was identical in the 2 treatment groups. At the end of the rt-PA infusion the circulating fibrinogen level was 61±35% of the starting value, as measured by a coagulationrate assay, and 69±25% as measured by sodium sulphite precipitation. After streptokinase infusion, corresponding fibrinogen levels were 12±18% and 20±11%. In the rt-PA group only 4·5% of the fibrinogen was measured as incoagulable fibrinogen degradation products, compared with 30% in the streptokinase group. Activation of the systemic fibrinolytic system was far less pronounced with rt-PA than with steptokinase. -
Introduction
fear of haemorrhagic complications and, less of pyrogenic and allergic reactions has limited extensive use of intravenous streptokinase as a thrombolytic agent in patients with acute myocardial infarction. This explains the keen interest in tissue-type plasminogen activator (t-PA), a naturally occurring protein in man which has a greater clot selectivity than streptokinase and urokinase and which, at thrombolytic doses, has not induced systemic fibrinolysis in animalsl-7 or in patients treated so far.8-10 The aim of this single-blind randomised trial in 7 centres was to compare the relative effectiveness, in terms of angiographically proven coronary patency, of 2 intravenous drug regimens-heparin plus human rt-PA versus heparin plus streptokinase. The safety of the 2 drug regimens was also compared, and their effect on components of the coagulation and fibrinolytic system assessed. The other 6 centres of the European Cooperative Study Group for Recombinant THE
seriously,
Schmutzler, W. Rutsch, M. Schartl); Hôpital Universitaire Saint-Pierre, Brussel (R. Bernard, M. De Marneffe, M. Renard, E. Van Thiel); Catharina Hospitaal, Eindhoven (H. R. Michels); Ziekenhuis Leyenburg, The Hague (G. A. Van de Kley); Städtische Kliniken, Fulda (W. Huhmann, G. Strupp, H. Nieth); 2. Medizinische Universitätsklinik, UK Eppendorf, Hamburg (W Bleifeld, D. Mathey, J. Schofer, K. H. Kuck, H. Becher); Universitair Ziekenhuis Gasthuisberg, Leuven (H. De Geest, F. Van de Werf, J Vanhaecke); II Medizinische Klinik und Poliklinik, Johannes GutenbergUniversität, Mainz (J. Meyer, R. Erbel, K. J. Henrichs, T. Pop); Centre Hospitalier Régional et Universitaire, Marseille (A. Serradimigni, M. Bory,J. Sainsous, M. Benichou); Hôpital Tenon, Paris (J. Acar, A. Vahanian, P. L Michel, J. M. Weber), Centre Hospitalier Régional et Universitaire, Tours (M. Brochier, Ph. Raynaud, B. Charbonnier).
843
Tissue-type Plasminogen Activator are concurrently comparing the effect of intravenously administered rt-PA and placebo on the patency rate of infarct-related vessels in patients with acute myocardial infarction.
TABLE I-SELECTED BASELINE CHARACTERISTICS OF PATIENTS ON
ADMISSION
Patients and Methods Patients 21 and 70 years old and without previous Patients myocardial infarotion were eligible for participation if severe chest pain typical of myocardial ischaemia persisted for at least 30 min. In addition, at least a 2 mm ST-segment elevation (60 ms after J point) in 2 or more standard frontal plane leads or 3 mm in 2 or more precordial leads had to be present. Presence of Q waves in the leads also showing ST-segment elevation did not exclude the patient from the trial. All patients were randomised within 6 h of the onset of pain. Patients were excluded if they met any of the following criteria: hypotension (systolic blood pressure below 90 mm Hg) and a heart-rate over 110/min in a clinical setting typical of cardiogenic shock; history of previous infarction, cerebrovascular accident, major surgery during the previous 6 months, or gastrointestinal bleeding during the previous 3 months; severe hypertension (over 200 mm Hg systolic pressure); prolonged or traumatic heart massage or artificial respiration; oral anticoagulation, a known bleeding disorder, or recent major trauma, particularly head injury; major hepatic or renal disease, cancer, or proliferative diabetic retinopathy; known alcohol or drug abuse; pregnancy or actual menstruation; anticipated problems with heart catheterisation, inability to cooperate, or anticipated problems with follow-up; previous coronary or non-coronary heart surgery. After obtaining informed consent the enrolling physician telephoned identification of an eligible patient to the data centre.
between
The data centre instructed the investigator which coded prepackaged vials, containing either rt-PA or streptokinase, were to be given. Allocation was at random and balanced at each clinic. Computations done in the design phase showed that if 20% of the angiograms of each group (rt-PA and streptokinase) were scored as non-perfused the 95% confidence interval of the difference would range from -15% to + 15%, assuming a total study size of 120 patients. Thus, we aimed for a total of 120 patients, 60 allocated to rt-PA and 60 to streptokinase.
Treatment
.
Patients randomised to the rt-PA group received a bolus injection of 5000 IU heparin followed by rt-PA (0-75 mg/kg body-weight) (Genentech Inc, supplied by Boehringer Ingelheim International GmbH) given by infusion pump over 90 min; the dose of rt-PA was based on the results ofaprevious trial. 11The rt-PA was produced by expression of the cloned human t-PA gene!2 in mammalian-cell culture. The patients allocated to streptokinase treatment received an intravenous bolus injection of 5000 IU heparin, 0-25 g methylprednisolone, and 0 -5gacetylsalicylic acid, followed by an infusion of 250 ml of 5% glucose containing 1 500 000 IU streptokinase (Hoechst-purchased from Behring Werke) given by infusion pump over 60 min; this dose was based on the experience of Schroder et al. 13 Blood samples were collected for enzyme and coagulation studies before the start of infusion. As soon as the intravenous treatments were started the patient was transferred to the cardiac catheterisation laboratory. Selective coronary arteriography was done between 75 min and 90 min after the start of the infusion of the thrombolytic drug. Individual centres were free to choose between the Sones and Judkins techniques and to decide which coronary vessel to inject first. At least 2 views were required of the right coronary artery and 3 of the left. Coronary angiography was recorded on 35 mm cine film, and this record as evaluated by the independent angiography evaluation group represented the primary end-point of the study. Once angiography had been done all subsequent diagnostic or therapeutic procedures were at the discretion of the individual centres. No antiarrhythmic drugs were given prophylactically, but they were used as required. Analgesics were given ad libitum. All medication used was recorded on the trial proforma.
MI=myocardial infarction,
BP blood pressure, BPM= beats per min.
Blood pressure and heart rate, clinical complications, and possible the treatment were followed during and after infusion while the patient was still in the coronary-care unit and during further hospital stay.
complications of
Assessment
of Coronary Arteriograms
angiogram was read by 2 assessors from a panel of experienced cardiologists and radiologists independent of the participating units. To maintain consistency one of them (D.P.dB.) was always present. The assessors knew the code number of the arteriogram and whether or not the electrocardiogram at the time of entry showed ST elevation in leads (V2, V3, and V4). They did not know which treatment the patient had received. Each assessor scored the arteriogram individually, segment by segment, and Each
recorded the results of visual assessment on a standard form with the of a predetermined qualitative code: (0) normal vessel; (1) mild stenosis less than 50% of the vessel diameter; (2) moderate stenosis greater than 50% but less than 90%; (3) severe stenosis greater than 90%, but distal vessels fill completely, not through collaterals, within 3 cardiac cycles; (4) subtotal occlusion, distal vessel does not fill within 3 cardiac cycles; (5) total occlusion with or without collateral filling. Collateral filling and left or right dominance were recorded, and the technical quality of the angiogram was assessed on a 10-point scale. In some angiograms a vessel filled poorly on the first injection of contrast, but subsequent injections showed better filling, either because of contrast-induced vasodilatation or because thrombus was displaced. These angiograms were scored on the basis of the first technically adequate injection. At the end of the session the 2 assessors compared their reports and reviewed any arteriograms that were discrepant. Pre-intervention angiography was not done in the present study, since we were anxious to start intravenous thrombolytic therapy as early as possible. The number of patients with initially patent or spontaneously reperfused vessels is thus unknown. Reperfusion in the context of the present trial is therefore a judgment based on the patency of vessels in the post-treatment angiogram, the appearance of vessels, in particular the presence of intraluminal thrombus, and knowledge of the presumed infarct-related vessel as determined by the electrocardiogram. use
844 Results
TABLE II-EVENTS BETWEEN START OF INFUSION AND END OF
CATHETERISATION
Between July 20 and Dec 31, 1984, 129 patients were randomised to the 2 treatment groups. The number of patients admitted in each trial centre were: Berlin 40, Aachen 32, Mainz 19, Eindhoven 13, Marseille 11, Fulda 8, Brussel 6.
Baseline characteristics are summarised in table 1. In all cases there was agreement between the treatment allocated and the results of rt-PA-antigen and fibrinolytic assays in plasma. In 1 patient the allotted streptokinase treatment could not be given because of shock; cardiac catheterisation was not attempted in this patient. Transmural infarction was confirmed in all patients; in over 97% this was based on a typical electrocardiogram pattern together with a typical increase of blood-enzyme levels. In the 128 patients who received treatment the full dose of rt-PA or streptokinase was given, and there was no difference between the 2 treatment blood pressure, heart rate, and groups in terms of allergic reactions (table II). Before catheterisation intravenous nitroglycerin was given in 66% of the rt-PA-treated patients and in 57% of the streptokinase-treated group. There were no deaths during the experimental infusion. Cardiac catheterisation could not be done in 1 patient because both
arrhythmia,
TABLE III-STATUS OF PRESUMED INFARCT-RELATED CORONARY ARTERY AS ASSESSED BY THE ANGIOGRAPHY EVALUATION GROUP
BP=blood pressure, BPM=beats per min, AV=atrioventricular, TIA=transient ischaemic attack, CVA = cerebrovascular accident, SK = streptokinase, PTCA percutaneous transluminal coronary angioplasty. =
Each assessor decided independently whether reperfusion had been successful, unsuccessful, or impossible to assess. Successful reperfusion had probably occurred if all the vessels in the infarctrelated area had scores of 3 or less and there were no "missing vessels". Reperfusion was assessed as unsuccessful if any of the vessels in the infarct-related area was seen as a totally occluded stump (code 5) or a subtotally occluded vessel with poor distal filling (code 4). 2% of angiograms could not be classified as reperfusion or non-reperfusion, in most cases because of technically inadequate
films.
Analysis of Blood Samples for Coagulation and Fibrinolytic Assays Blood samples collected in tubes for analysis in the Central Coagulation Laboratory were obtained within the 30 min before infusion and 60 min and 90 min after the start of the infusion. 2 tubes for blood collection were provided, 1 containing 0’ 5 ml sodium citrate (final concentration 0’ 01mol) and the other, citrate and aprotinin (final concentration 150 KIU/ml) to counteract proteolysis generated by plasmin in vitro. 4-55 ml blood was collected in each tube, centrifuged within 1 h, and the plasma stored at -20°C. Fibrinogen was determined by means of a clotting-rate assayl4 and sodium sulphite precipitation. 15 Fibrin degradation products 16 and rt-PA-antigen were also measured. 17 All data were collected on standardised forms which contained relevant procedural instructions. The chi-square test was used when appropriate and two-sided p values are given. 18 Central analysis was done independently from the investigators and Boehringer Ingelheim International. 95% confidence intervals of the difference between proportions were calculated with the method of Thomas and Gart.19 Continuous data are represented as medians and ranges; otherwise standard statistical tests were used.
was graded stenotic by the observer but was not available for review. t2 patients in this group did not have angiography for technical reasons and 1 film was graded stenotic by the observer but was not available for review.
* film
TABLE IV-EVENTS AFTER ADMINISTRATION OF INFUSION BUT
BEFORE HOSPITAL DISCHARGE
845
Top: Fibrinogen levels before, at 60, and at 90 min after start of the infusion with rt-PA (left) or streptokinase (right). Bottom: Levels offibrinogen degradation products before, at 60, and at 90 min after the start of the infusion with rt-PA (left) or streptokinase (right). Data represent individual
values; horizontal bar indicates the median.
femoral arteries were severely stenotic, but the data pertaining to this patient were included in the analysis. No excessive bleeding at the arterial puncture site was noted during catheterisation. There was 1 transient ischaemic attack in the rt-PA group and a confirmed cerebrovascular accident in the streptokinase group; neither event prevented
completion
of the
protocol.
Angiographic Findings Table III shows the numbers of patients with occluded or subtotally occluded infarct-related vessels and patients with
patent vessels in each treatment group. There is a 15% difference between the groups in favour of the rt-PA
(70% in the rt-PA-treated group and 55% in the streptokinase-treated group) (95% confidence interval, +32% to -207o; p=0054). Events during Hospital Follow-up Minor haemorrhagic complications during infusion, mainly bleeding at puncture sites, were rare, and frequency did not differ between the 2 treatment groups (table IV). No blood transfusion had to be given during the brief period of infusion and heart catheterisation. During the first 48 h after transfer to the coronary-care unit, haematoma and prolonged bleeding at puncture sites were more frequently a problem in the streptokinase-treated patients, but blood transfusions were given infrequently in both groups (table IV). treatment
846
Retroperitoneal bleeding, cerebral haemorrhage, melaena, and haematemesis were not reported in either treatment group. Only 1 patient had an allergic reaction; he had been treated with streptokinase. There was 1 transient ischaemic episode during catheterisation in an rt-PA-treated patient and a cerebrovascular accident in a streptokinase-treated patient who was discharged with a neurological deficit. Coagulation and Fibrinolytic Assays Infusion of rt-PA resulted in peak plasma levels of 1-30±0-74 g/ml (mean±SD) at 60 min. (These concentrations were determined by calibration against the International Reference Preparation for t-PA [83/517]. The present levels are a factor of 2 lower than would have been obtained by comparison with our previously used homereference preparation.) This large SD indicates large interindividual variability in response to a standard rate of infusion. Infusion of rt-PA resulted in a fall of the plasma fibrinogen level to 76±29% (median from 2 -5to 1’8) at 60 min and to 61+35% at 90 min (median 1’3) as measured with a clotting rate assay on frozen plasma samples collected with aprotinin (figure, top). The plasma fibrinogen level was below 1 g/1 in 1 out of 59 patients (2%) at the start of the infusion, in 8 at 60 min (13%), and in 17 at 90 min (28%). A fall below 0 -5g/l was observed in 1 out of 60 patients after 60 min and in 3 after 90 min. A smaller drop in the fibrinogen level was observed when measured with sodium sulphite precipitation: to 78±28% at 60 min and to 69±25% at 90 min (n = 36). The level of fibrinogen degradation products in serum rose in rt-PA-treated patients to 2 -2±2’7% (mean±SD, n= 54) of the baseline fibrinogen level after 60 min (n=46) and to 4 -6:!:5’ 20/0 at 90 min (figure, bottom). Streptokinase infusion resulted in extensive systemic fibrinolytic activation as shown by a fall in fibrinogen to 12±18% (median from 2-4 to 0 - 15) after 60 min and to 8±11% after 90 min (median 0’08), with a corresponding increase of serum fibrinogen degradation products representing up to 30% of the preinfusion value of fibrinogen. With the precipitation method the fibrinogen level was 20±11% of the starting value at the end of the infusion. In the streptokinase-treated group fibrinogen fell below 0 -55 gull in 54 of 62 patients (87%) at 90 min. Discussion The major drawback of streptokinase and urokinase is that they activate both fibrin-bound plasminogen and circulating plasminogen indiscriminately. The plasmin generated in the circulation is rapidly neutralised by the vast pool of a2-antiplasmin and is therefore lost for thrombolysis. Once the reserve of a2-antiplasmin has been exhausted, free circulating plasmin will degrade fibrinogen and other plasma proteins as well as fibrin, resulting in both a systemic fibrinolytic and a local thrombolytic state.21The danger of fibrinogen depletion is bleeding. The frequency of major haemorrhagic bleeding complications (requiring blood transfusion) in patients treated for myocardial infarction with streptokinase is less than 5% and of life-threatening haemorrhages less than 1 -5%.22 If thrombolysis is to have a therapeutic role in acute myocardial infarction, the intravenous route of drug administration is the only realistic one.23 The dose regimen of rt-PA was selected on the only therapeutic experience available when this trial was plannedland compared with a
of infusion intravenous short-term no drawback of The pretreatment having streptokinase.13 coronary angiography is that visualisation of the coronary circulation after thrombolytic treatment does not allow us to establish definite reperfusion but only to compare patency rate between 2 treatment groups. With the very strict myocardial infarction criteria applied in this trial and the short duration of clinical symptoms it is likely that 80% or more of the patients at admission had an occluded vessel in the infarct area.22,24 We chose to exclude vessels with poor distal flow from our definition of patency because clinical experience suggests that inadequate distal flow is irrelevant to
high-dose,
3
myocardial salvage.’
The 2 drug regimens were well tolerated. In 12 patients in whom the infarct-related coronary artery was still occluded after rt-PA, percutaneous transluminal coronary angioplasty (PTCA) with or without intracoronary streptokinase was carried out; 8 patients improved. Of 13 patients in whom the infarct-related artery was still occluded after systemic treatment with streptokinase, 11 had reperfusion after PTCA with or without intracoronary streptokinase. After the clinical end-point intracoronary streptokinase treatment was administered exclusively in 4 patients, of whom only 2 improved. These results are in line with an earlier study in which 6 patients unsuccessfully treated with systemic rt-PA remained completely unresponsive to intracoronary streptokinase.’1 It is not possible to analyse whether reocclusion occurred more frequently and rapidly with 1 of the 2 treatments compared in this trial since the end-point was set at the coronary angiography obtained 90 min after the start of thrombolytic treatment and further treatment was left to the discretion of the investigator. The slightly higher incidence of arrhythmia and atrioventricular block in the rt-PA-treated patients may be related to a higher reperfusion rate. 25 Various degrees of atrioventricular block are associated with reperfusion in the inferoposterior myocardium.26 The early bolus of 5000 IU of heparin together with mechanical trauma may have caused the bleeding, which occurred mainly at the arterial catheterisation site. There were appreciably fewer bleeding events in the rt-PA group. As expected, streptokinase treatment induced an intense activation of the circulating fibrinolytic system with a circulating fibrinogen level of 12% (clotting-rate assay) or 20% (precipitation method) of the starting value at the end of the infusion. With 0’ 75 mg rt-PA/kg body-weight the circulating fibrinogen at the end of the infusion fell to 61 % (clotting-rate assay) or 69% (precipitation method) of the value before infusion. A remarkable discrepancy was observed between the extent of fibrinogen degradation and the appearance of degradation products. A fall in fibrinogen of 29% at 90 min in the rt-PA group corresponded to a recovery of only 4-5% as degradation products, whereas a fall in fibrinogen of 88% in the streptokinase group was associated with 30% recovery as coagulable fibrinogen degradation products. This finding suggests that fibrinogen breakdown during rt-PA infusion is less extensive, yielding molecules which are not measured in the fibrinogen assays but are nevertheless incorporated in the clot when serum is obtained for the assay of fibrinogen
degradation products. In this trial treatment with rt-PA was associated with a higher patency rate of the infarct-related vessel; the confidence interval of the difference does not exclude the possibility that rt-PA and streptokinase are in fact of equal
847
efficacy but rt-PA treatment produced considerably systemic fibrinolysis.
less
We thank Dr W. Feuerer and the staff of the Dr Karl Thomae GmbH for
help with the preparation and dispatch of the clinical trial supplies. Correspondents should be addressed to M. V. Center for Thrombosis and Vascular Research, Campus Gasthuisberg, K.U.Leuven, Herestraat 49, B-3000 Leuven, Belgium.
BE, Geltman EM, Tiefenbrunn AJ, et al. Improvement of regional myocardial metabolism after coronary thrombolysis induced with tissue-type plasminogen activator or streptokinase. Circulation. 1984; 69: 983-90. 11. Collen D, Topol EJ, Tiefenbrunn AJ, et al. Coronary thrombolysis with recombinant human tissue-type plasminogen activator a prospective, randomized, placebocontrolled trial. Circulation 1984; 70: 1012-17. 12. Pennica D, Holmes WR, Kohr WJ, et al. Cloning and expression of human tissue-type plasminogen activator cDNA in E coli. Nature 1983; 301: 214-21. 13. Schröder R, Biamino G, von Leitner ER, et al. Systemische Thrombolyse mit Streptokinase-Kurzzeitinfusion bei akuten Myokardinfarkt. Z Kardiol 1982; 71:
10. Sobel
709-18.
Vermylen C, De Vreker RA, Verstraete M. A rapid enzymatic method for the assay of fibrinogen: the fibrin polymerization test (FPT). Clin Chim Acta 1983; 8: 418-24 Rampling MW, Gaffney PJ. The sulphite precipitation method for fibrinogen measurement; its use on small samples in the presence of fibrinogen degradation
14.
REFERENCES 1 Matsuo
15.
O, Rijken DC, Collen D. Thrombolysis by human
tissue
plasminogen
activator and urokinase in rabbits with experimental pulmonary embolus. Nature 1981; 291: 590-91. 2 Korninger C, Matsuo O, Suy R, Stassen JM, Collen D. Thrombolysis with human extrinsic (tissue-type) plasminogen activator in dogs with femoral vein thrombosis. J Clin Invest 1982, 69: 573-80. 3. Collen D, Stassen JM, Verstraete M. Thrombolysis with human extrinsic (tissue-type) plasminogen activator in rabbits with experimental jugular vein thrombosis. Effect of molecular form and dose of activator, age of the thrombus and route of administration. J Clin Invest 1983; 71: 368-76. 4. Bergmann SR, Fox KAA, Ter-Pogossian MM, Sobel BE, Collen D. Clot-selective coronary thrombolysis with tissue-type plasminogen activator. Science 1983; 220: 1181-83. 5 Van de Werf F, Bergmann SR, Fox KAA, et al. Coronary thrombolysis with intravenously administered human tissue-type plasminogen activator produced by 6 7
8
9.
recombinant DNA technology. Circulation 1984; 69: 605-10. Gold HK, Fallon JT, Yasuda T, et al. Coronary thrombolysis with recombinant human tissue-type plasminogen activator. Circulation 1984; 70: 700-07. Flameng W, Van de Werf F, Vanhaecke J, Verstraete M, Collen D. Coronary thrombolysis and infarct size reduction after intravenous infusion of recombinant tissue-type plasminogen activator in nonhuman primates. J Clin Invest 1985; 75: 84-90. Weimar W, Stibbe J, van Seyen A, Billiau P, De Somer P, Collen D. Specific lysis of an iliofemoral thrombus by administration of extrinsic (tissue-type) plasminogen activator. Lancet 1981; ii: 1018-20. Van de Werf F, Ludbrook PA, Bergmann SR, et al. Coronary thrombolysis with tissuetype plasminogen activator in patients with evolving myocardial infarction. N Engl J Med 1984, 310: 609-13.
Hypothesis OESTROGEN DEFICIENCY AFTER TUBAL LIGATION
JOHN CATTANACH 140 Power Street, Hawthorn
3122, Australia
who had undergone tubal within the past seven years were found to have oestrogen excretion concentrations at ovulation below the tenth percentile. A disturbance in the oestrogen/progesterone ratio as a consequence of localised hypertension at the ovary, when the utero-ovarian arterial loop is occluded at tubal ligation, is proposed as a possible cause of oestrogen deficiency syndrome, dysfunctional uterine bleeding, and menorrhagia after tubal ligation. Similar pathophysiology may occur after hysterectomy with ovarian conservation.
Summary
4 of 7
women
ligation
products. Clin Chim Acta 1976;
67: 43-52.
16
Merskey C, Lalezari P, Johnson AJ. A rapid simple sensitive method for measuring fibrinolytic split products in human serum. Proc Soc Exp Biol Med 1969; 131:
17.
Rijken DC, Juhan-Vague I, De Cock F, Collen D. Measurement of human tissue-type plasminogen activator by a two-site immunoradiometric assay. J Lab Clin Med 1983;
871-78.
101: 274-94. 18. Snedecor GW, Cochran WG. Statistical methods. Iowa: Iowa State
University Press,
1973: 114-16. Thomas DG, Gart JJ. A table of exact confidence limits for differences and ratios of two proportions and their odds ratios. J Am Stat Soc 1977; 72: 73-76. Gaffney PJ, Curtis AD. A collaborative study of a proposed international standard for tissue plasminogen activator (t-PA). Thromb Haemostas 1985; 53: 134-36 Collen D On the regulation and control of fibrinolysis. Edward Kowalski Memorial Lecture Thromb Haemostas 1980; 43: 77-89. Spann JF, Sherry S. Coronary thrombolysis for evolving myocardial infarction. Drugs
19.
20. 21. 22.
1984; 28: 465-83. 23. Verstraete M Intravenous administration ofa 24. 25.
26
thrombolytic agent is the only realistic therapeutic approach in evolving myocardial infarction. Eur Heart J1985 (in press). Laffel GL, Braunwald E. A new strategy for the treatment of acute myocardial infarction. N Engl J Med 1984; 311: 710-17; 770-76. Goldberg S, Greenspon AJ, Urban PL, et al. Reperfusion arrhythmia. a marker of restoration of integrated flow during intracoronary thrombolysis for acute myocardial infarction. Am Heart J 1983; 105: 26-32. Wei JY, Markis JE, Malagold M, Braunwald E. Cardiovascular reflexes stimulated by reperfusion of ischemic myocardium in acute myocardial infarction. Circulation 1983;
67: 796-801.
PATIENTS AND METHODS
Between July, 1975, and July, 1977, 601 women were instructed in the ovulation method with optional use of a diaphragm.9 They were followed up by a postal questionnaire in December, 1982. Women who had learned the ovulation method and had subsequently undergone tubal ligation were contacted. The best time to assess oestrogen status is at ovulation (Brown JB, personal
communication). To qualify for
inclusion the women had to be able to identify and the peak symptom (the last day on which cervical mucus has a distinctly lubricative quality) and be prepared to collect urine in 24-h specimen lots on the day before, the day of, and a day after identification of ovulation. Two extra women who had under-gone tubal ligation within the time limits and fitted the inclusion criteria, and had no menstrual irregularities, were also studied. Urine samples were analysed for total oestrogen excretion and associated pregnanediol in Professor E. J. Brown’s biochemistry laboratory at Melbourne University. fertile
mucus
RESULTS
13 women were eligible. 7 agreed to cooperate. In 4 of these total oestrogen levels on the day of ovulation were below the TABLE I-PUBLISHED REPORTS ON
SEQUELAE OF
TUBAL LIGATION
INTRODUCTION
No
satisfactory explanation appears to have been established for the reportedly increased incidence of menorrhagia, abnormal uterine bleeding, and major surgery after tubal ligationl-7 (table I). Since an underlying disturbance in the oestrogen/progesterone ratio is a likely explanation for these effects, I report the results of measuring concentrations of these hormones at ovulation in women who were able to use the Billings ovulation method8 to identify ovulation and who had subsequently undergone tubal ligation.
TL=tubal Available.
ligation.
Retro Retrospective.
Pros Prospective.
NA=Not
in a patient who had had 2 previous complications. The fact that the study was not double blind is also unlikely to have biased the results, although it may have influenced the frequency of caesarean sections. Unfortunately, we did not measure platelet aggregation in mothers, so we cannot be sure whether they complied with the drug regimen. Also, we did not check on whether women in the control group took aspirin during the pregnancy. The combination of dipyridamole and aspirin did not cause apparent harm to mother or fetus, although the safety of aspirin during pregnancy has been seriously questioned by workers who have reported a high frequency of abnormal bleeding at delivery and/or of mucosal or cutaneous haemorrhage in babies. 10 The absence of such complications in our series, although they were carefully looked for, suggests that the doses of aspirin we gave, which is lower than those usually given, are likely to be harmless to the pregnant mother and her baby.
RANDOMISED TRIAL OF INTRAVENOUS RECOMBINANT TISSUE-TYPE PLASMINOGEN ACTIVATOR VERSUS INTRAVENOUS STREPTOKINASE IN ACUTE MYOCARDIAL
INFARCTION
Report from the European Cooperative Study Group for Recombinant Tissue-type Plasminogen Activator* M. VERSTRAETE M. BORY D. COLLEN R. ERBEL
R. J. LENNANE D. MATHEY H. R. MICHELS M. SCHARTL R. UEBIS
Summary
R. BERNARD R. W. BROWER D. P. DE BONO W. HUHMANN J. LUBSEN J. MEYER W. RUTSCH W. SCHMIDT R. VON ESSEN
In a single-blind randomised trial in patients with acute myocardial infarction of less than
6 h duration, the frequency of coronary patency was found to be higher after intravenous administration of recombinant human tissue-type plasminogen activator (rt-PA) than after intravenous streptokinase. 64 patients were allocated to 0·75 mg rt-PA/kg over 90 min, and the infarct-related coronary artery was patent in 70% of 61 assessable coronary angiograms taken 75-90 min after the start of infusion; 65 patients were allocated to 1 500 000 IU streptokinase over 60 min, and the infarct-related vessel was patent in 55% of 62 assessable angiograms. The 95% confidence interval of the difference ranges from ±30 to -2% (p = 0·054). Bleeding *STEERING COMMITTEE: M. Verstraete (Leuven) (chairman); R. J. Lennane (Ingelheim) (clinical coordinator); D. P. de Bono (Edinburgh); J. Lubsen (Rotterdam); D. Mathey (Hamburg); R. von Essen (Aachen). ADVISORY BOARD: J. Hampton (Nottingham); H. J. Jesdinsky (Düsseldorf); D. G. Julian (Newcastle upon Tyne); W. Schaper (Bad Nauheim); L. Wilhelmsen (Göteborg). DATA CENTRE: J. Lubsen, R. W. Brower, P. Fioretti, B. Soward, M. Bokslag (Thoraxcentre, Rotterdam). ANGIOGRAPHY EVALUATION GROUP: D. P. de Bono (Edinburgh) (secretary); W. S. Hillis (Glasgow); D. Reid (Newcastle); C. Turnbull (Edinburgh). CENTRAL COAGULATION LABORATORY: D. Collen, H. R. Lijnen (Leuven). PARTICIPATING CLINICAL CENTRES: Innere Medizin I, RheinischWestfälische Technische Hochschule, Aachen (S. Effert, R. von Essen, R. Uebis, W. Schmidt); Academisch Medisch Centrum, Amsterdam (A. J. Dunning, K. Romyn); Klinikum Charlottenburg, Freie Universität Berlin (H.
Correspondence should be addressed to M. B., Service de Néphrologle, Hôpital Tenon, 4 Rue de la Chine, 75970 Paris, Cedex 20, France. REFERENCES
propos du poids et de la taille des nouveaux nés à la naissance. Rev Franc Gynec 1971; 66: 391-96. 2. Bonnar J, Redman CWG, Denson KW. The role of coagulation and fibrinolysis in preeclampsia. In- Lindheimer MD, Katz AI, Zuspan F, eds. Hypertension in pregnancy. New York: Wiley, 1976: 85-93. 3. Redman CWG, Bonner J, Beilin L. Early platelet consumption in pre-eclampsia Lancet 1978; i: 467-69. 4. McKay DG. Chronic intravascular coagulation in normal pregnancy and preeclampsia. Contributions Nephrol 1981; 25: 108-19. 5. Howie PW, Prentice CRM, Forbes CD. Failure of heparin therapy to affect the clinical course of severe preeclampsia. Br J Obstet Gynaecol 1979; 82: 711-17 6. Marcus AJ. Editorial retrospective: Aspirin as an antithrombotic medication N Engl J Med 1983; 309: 1515-17. 7 Salzman EW. Aspirin to prevent arterial thrombosis. N Engl J Med 1982; 307: 113-15. 8. Lewis HD, Davis JW, Archibald DG, et al. Protective effects of aspirin against acute myocardial infarction and death in men with unstable angina. Results of a Veterans Administration cooperative study. N Engl J Med 1983; 309: 396-403. 9. Persantin-Aspirin Reinfarction Study Research Group. Persantin and aspirin in coronary heart disease. Circulation 1980; 62: 449-61. 10 Stuart MJ, Gross SJ, Elrad H, Graeber JE. Effects of acetylsalicylic acid ingestion on maternal and neonatal hemostasis. N Engl J Med 1982; 307: 909-12. 1.
Leroy B, Lefort F. A
episodes and other complications were less common in the rt-PA patients than in the streptokinase group. Hospital mortality was identical in the 2 treatment groups. At the end of the rt-PA infusion the circulating fibrinogen level was 61±35% of the starting value, as measured by a coagulationrate assay, and 69±25% as measured by sodium sulphite precipitation. After streptokinase infusion, corresponding fibrinogen levels were 12±18% and 20±11%. In the rt-PA group only 4·5% of the fibrinogen was measured as incoagulable fibrinogen degradation products, compared with 30% in the streptokinase group. Activation of the systemic fibrinolytic system was far less pronounced with rt-PA than with steptokinase. -
Introduction
fear of haemorrhagic complications and, less of pyrogenic and allergic reactions has limited extensive use of intravenous streptokinase as a thrombolytic agent in patients with acute myocardial infarction. This explains the keen interest in tissue-type plasminogen activator (t-PA), a naturally occurring protein in man which has a greater clot selectivity than streptokinase and urokinase and which, at thrombolytic doses, has not induced systemic fibrinolysis in animalsl-7 or in patients treated so far.8-10 The aim of this single-blind randomised trial in 7 centres was to compare the relative effectiveness, in terms of angiographically proven coronary patency, of 2 intravenous drug regimens-heparin plus human rt-PA versus heparin plus streptokinase. The safety of the 2 drug regimens was also compared, and their effect on components of the coagulation and fibrinolytic system assessed. The other 6 centres of the European Cooperative Study Group for Recombinant THE
seriously,
Schmutzler, W. Rutsch, M. Schartl); Hôpital Universitaire Saint-Pierre, Brussel (R. Bernard, M. De Marneffe, M. Renard, E. Van Thiel); Catharina Hospitaal, Eindhoven (H. R. Michels); Ziekenhuis Leyenburg, The Hague (G. A. Van de Kley); Städtische Kliniken, Fulda (W. Huhmann, G. Strupp, H. Nieth); 2. Medizinische Universitätsklinik, UK Eppendorf, Hamburg (W Bleifeld, D. Mathey, J. Schofer, K. H. Kuck, H. Becher); Universitair Ziekenhuis Gasthuisberg, Leuven (H. De Geest, F. Van de Werf, J Vanhaecke); II Medizinische Klinik und Poliklinik, Johannes GutenbergUniversität, Mainz (J. Meyer, R. Erbel, K. J. Henrichs, T. Pop); Centre Hospitalier Régional et Universitaire, Marseille (A. Serradimigni, M. Bory,J. Sainsous, M. Benichou); Hôpital Tenon, Paris (J. Acar, A. Vahanian, P. L Michel, J. M. Weber), Centre Hospitalier Régional et Universitaire, Tours (M. Brochier, Ph. Raynaud, B. Charbonnier).
843
Tissue-type Plasminogen Activator are concurrently comparing the effect of intravenously administered rt-PA and placebo on the patency rate of infarct-related vessels in patients with acute myocardial infarction.
TABLE I-SELECTED BASELINE CHARACTERISTICS OF PATIENTS ON
ADMISSION
Patients and Methods Patients 21 and 70 years old and without previous Patients myocardial infarotion were eligible for participation if severe chest pain typical of myocardial ischaemia persisted for at least 30 min. In addition, at least a 2 mm ST-segment elevation (60 ms after J point) in 2 or more standard frontal plane leads or 3 mm in 2 or more precordial leads had to be present. Presence of Q waves in the leads also showing ST-segment elevation did not exclude the patient from the trial. All patients were randomised within 6 h of the onset of pain. Patients were excluded if they met any of the following criteria: hypotension (systolic blood pressure below 90 mm Hg) and a heart-rate over 110/min in a clinical setting typical of cardiogenic shock; history of previous infarction, cerebrovascular accident, major surgery during the previous 6 months, or gastrointestinal bleeding during the previous 3 months; severe hypertension (over 200 mm Hg systolic pressure); prolonged or traumatic heart massage or artificial respiration; oral anticoagulation, a known bleeding disorder, or recent major trauma, particularly head injury; major hepatic or renal disease, cancer, or proliferative diabetic retinopathy; known alcohol or drug abuse; pregnancy or actual menstruation; anticipated problems with heart catheterisation, inability to cooperate, or anticipated problems with follow-up; previous coronary or non-coronary heart surgery. After obtaining informed consent the enrolling physician telephoned identification of an eligible patient to the data centre.
between
The data centre instructed the investigator which coded prepackaged vials, containing either rt-PA or streptokinase, were to be given. Allocation was at random and balanced at each clinic. Computations done in the design phase showed that if 20% of the angiograms of each group (rt-PA and streptokinase) were scored as non-perfused the 95% confidence interval of the difference would range from -15% to + 15%, assuming a total study size of 120 patients. Thus, we aimed for a total of 120 patients, 60 allocated to rt-PA and 60 to streptokinase.
Treatment
.
Patients randomised to the rt-PA group received a bolus injection of 5000 IU heparin followed by rt-PA (0-75 mg/kg body-weight) (Genentech Inc, supplied by Boehringer Ingelheim International GmbH) given by infusion pump over 90 min; the dose of rt-PA was based on the results ofaprevious trial. 11The rt-PA was produced by expression of the cloned human t-PA gene!2 in mammalian-cell culture. The patients allocated to streptokinase treatment received an intravenous bolus injection of 5000 IU heparin, 0-25 g methylprednisolone, and 0 -5gacetylsalicylic acid, followed by an infusion of 250 ml of 5% glucose containing 1 500 000 IU streptokinase (Hoechst-purchased from Behring Werke) given by infusion pump over 60 min; this dose was based on the experience of Schroder et al. 13 Blood samples were collected for enzyme and coagulation studies before the start of infusion. As soon as the intravenous treatments were started the patient was transferred to the cardiac catheterisation laboratory. Selective coronary arteriography was done between 75 min and 90 min after the start of the infusion of the thrombolytic drug. Individual centres were free to choose between the Sones and Judkins techniques and to decide which coronary vessel to inject first. At least 2 views were required of the right coronary artery and 3 of the left. Coronary angiography was recorded on 35 mm cine film, and this record as evaluated by the independent angiography evaluation group represented the primary end-point of the study. Once angiography had been done all subsequent diagnostic or therapeutic procedures were at the discretion of the individual centres. No antiarrhythmic drugs were given prophylactically, but they were used as required. Analgesics were given ad libitum. All medication used was recorded on the trial proforma.
MI=myocardial infarction,
BP blood pressure, BPM= beats per min.
Blood pressure and heart rate, clinical complications, and possible the treatment were followed during and after infusion while the patient was still in the coronary-care unit and during further hospital stay.
complications of
Assessment
of Coronary Arteriograms
angiogram was read by 2 assessors from a panel of experienced cardiologists and radiologists independent of the participating units. To maintain consistency one of them (D.P.dB.) was always present. The assessors knew the code number of the arteriogram and whether or not the electrocardiogram at the time of entry showed ST elevation in leads (V2, V3, and V4). They did not know which treatment the patient had received. Each assessor scored the arteriogram individually, segment by segment, and Each
recorded the results of visual assessment on a standard form with the of a predetermined qualitative code: (0) normal vessel; (1) mild stenosis less than 50% of the vessel diameter; (2) moderate stenosis greater than 50% but less than 90%; (3) severe stenosis greater than 90%, but distal vessels fill completely, not through collaterals, within 3 cardiac cycles; (4) subtotal occlusion, distal vessel does not fill within 3 cardiac cycles; (5) total occlusion with or without collateral filling. Collateral filling and left or right dominance were recorded, and the technical quality of the angiogram was assessed on a 10-point scale. In some angiograms a vessel filled poorly on the first injection of contrast, but subsequent injections showed better filling, either because of contrast-induced vasodilatation or because thrombus was displaced. These angiograms were scored on the basis of the first technically adequate injection. At the end of the session the 2 assessors compared their reports and reviewed any arteriograms that were discrepant. Pre-intervention angiography was not done in the present study, since we were anxious to start intravenous thrombolytic therapy as early as possible. The number of patients with initially patent or spontaneously reperfused vessels is thus unknown. Reperfusion in the context of the present trial is therefore a judgment based on the patency of vessels in the post-treatment angiogram, the appearance of vessels, in particular the presence of intraluminal thrombus, and knowledge of the presumed infarct-related vessel as determined by the electrocardiogram. use
844 Results
TABLE II-EVENTS BETWEEN START OF INFUSION AND END OF
CATHETERISATION
Between July 20 and Dec 31, 1984, 129 patients were randomised to the 2 treatment groups. The number of patients admitted in each trial centre were: Berlin 40, Aachen 32, Mainz 19, Eindhoven 13, Marseille 11, Fulda 8, Brussel 6.
Baseline characteristics are summarised in table 1. In all cases there was agreement between the treatment allocated and the results of rt-PA-antigen and fibrinolytic assays in plasma. In 1 patient the allotted streptokinase treatment could not be given because of shock; cardiac catheterisation was not attempted in this patient. Transmural infarction was confirmed in all patients; in over 97% this was based on a typical electrocardiogram pattern together with a typical increase of blood-enzyme levels. In the 128 patients who received treatment the full dose of rt-PA or streptokinase was given, and there was no difference between the 2 treatment blood pressure, heart rate, and groups in terms of allergic reactions (table II). Before catheterisation intravenous nitroglycerin was given in 66% of the rt-PA-treated patients and in 57% of the streptokinase-treated group. There were no deaths during the experimental infusion. Cardiac catheterisation could not be done in 1 patient because both
arrhythmia,
TABLE III-STATUS OF PRESUMED INFARCT-RELATED CORONARY ARTERY AS ASSESSED BY THE ANGIOGRAPHY EVALUATION GROUP
BP=blood pressure, BPM=beats per min, AV=atrioventricular, TIA=transient ischaemic attack, CVA = cerebrovascular accident, SK = streptokinase, PTCA percutaneous transluminal coronary angioplasty. =
Each assessor decided independently whether reperfusion had been successful, unsuccessful, or impossible to assess. Successful reperfusion had probably occurred if all the vessels in the infarctrelated area had scores of 3 or less and there were no "missing vessels". Reperfusion was assessed as unsuccessful if any of the vessels in the infarct-related area was seen as a totally occluded stump (code 5) or a subtotally occluded vessel with poor distal filling (code 4). 2% of angiograms could not be classified as reperfusion or non-reperfusion, in most cases because of technically inadequate
films.
Analysis of Blood Samples for Coagulation and Fibrinolytic Assays Blood samples collected in tubes for analysis in the Central Coagulation Laboratory were obtained within the 30 min before infusion and 60 min and 90 min after the start of the infusion. 2 tubes for blood collection were provided, 1 containing 0’ 5 ml sodium citrate (final concentration 0’ 01mol) and the other, citrate and aprotinin (final concentration 150 KIU/ml) to counteract proteolysis generated by plasmin in vitro. 4-55 ml blood was collected in each tube, centrifuged within 1 h, and the plasma stored at -20°C. Fibrinogen was determined by means of a clotting-rate assayl4 and sodium sulphite precipitation. 15 Fibrin degradation products 16 and rt-PA-antigen were also measured. 17 All data were collected on standardised forms which contained relevant procedural instructions. The chi-square test was used when appropriate and two-sided p values are given. 18 Central analysis was done independently from the investigators and Boehringer Ingelheim International. 95% confidence intervals of the difference between proportions were calculated with the method of Thomas and Gart.19 Continuous data are represented as medians and ranges; otherwise standard statistical tests were used.
was graded stenotic by the observer but was not available for review. t2 patients in this group did not have angiography for technical reasons and 1 film was graded stenotic by the observer but was not available for review.
* film
TABLE IV-EVENTS AFTER ADMINISTRATION OF INFUSION BUT
BEFORE HOSPITAL DISCHARGE
845
Top: Fibrinogen levels before, at 60, and at 90 min after start of the infusion with rt-PA (left) or streptokinase (right). Bottom: Levels offibrinogen degradation products before, at 60, and at 90 min after the start of the infusion with rt-PA (left) or streptokinase (right). Data represent individual
values; horizontal bar indicates the median.
femoral arteries were severely stenotic, but the data pertaining to this patient were included in the analysis. No excessive bleeding at the arterial puncture site was noted during catheterisation. There was 1 transient ischaemic attack in the rt-PA group and a confirmed cerebrovascular accident in the streptokinase group; neither event prevented
completion
of the
protocol.
Angiographic Findings Table III shows the numbers of patients with occluded or subtotally occluded infarct-related vessels and patients with
patent vessels in each treatment group. There is a 15% difference between the groups in favour of the rt-PA
(70% in the rt-PA-treated group and 55% in the streptokinase-treated group) (95% confidence interval, +32% to -207o; p=0054). Events during Hospital Follow-up Minor haemorrhagic complications during infusion, mainly bleeding at puncture sites, were rare, and frequency did not differ between the 2 treatment groups (table IV). No blood transfusion had to be given during the brief period of infusion and heart catheterisation. During the first 48 h after transfer to the coronary-care unit, haematoma and prolonged bleeding at puncture sites were more frequently a problem in the streptokinase-treated patients, but blood transfusions were given infrequently in both groups (table IV). treatment
846
Retroperitoneal bleeding, cerebral haemorrhage, melaena, and haematemesis were not reported in either treatment group. Only 1 patient had an allergic reaction; he had been treated with streptokinase. There was 1 transient ischaemic episode during catheterisation in an rt-PA-treated patient and a cerebrovascular accident in a streptokinase-treated patient who was discharged with a neurological deficit. Coagulation and Fibrinolytic Assays Infusion of rt-PA resulted in peak plasma levels of 1-30±0-74 g/ml (mean±SD) at 60 min. (These concentrations were determined by calibration against the International Reference Preparation for t-PA [83/517]. The present levels are a factor of 2 lower than would have been obtained by comparison with our previously used homereference preparation.) This large SD indicates large interindividual variability in response to a standard rate of infusion. Infusion of rt-PA resulted in a fall of the plasma fibrinogen level to 76±29% (median from 2 -5to 1’8) at 60 min and to 61+35% at 90 min (median 1’3) as measured with a clotting rate assay on frozen plasma samples collected with aprotinin (figure, top). The plasma fibrinogen level was below 1 g/1 in 1 out of 59 patients (2%) at the start of the infusion, in 8 at 60 min (13%), and in 17 at 90 min (28%). A fall below 0 -5g/l was observed in 1 out of 60 patients after 60 min and in 3 after 90 min. A smaller drop in the fibrinogen level was observed when measured with sodium sulphite precipitation: to 78±28% at 60 min and to 69±25% at 90 min (n = 36). The level of fibrinogen degradation products in serum rose in rt-PA-treated patients to 2 -2±2’7% (mean±SD, n= 54) of the baseline fibrinogen level after 60 min (n=46) and to 4 -6:!:5’ 20/0 at 90 min (figure, bottom). Streptokinase infusion resulted in extensive systemic fibrinolytic activation as shown by a fall in fibrinogen to 12±18% (median from 2-4 to 0 - 15) after 60 min and to 8±11% after 90 min (median 0’08), with a corresponding increase of serum fibrinogen degradation products representing up to 30% of the preinfusion value of fibrinogen. With the precipitation method the fibrinogen level was 20±11% of the starting value at the end of the infusion. In the streptokinase-treated group fibrinogen fell below 0 -55 gull in 54 of 62 patients (87%) at 90 min. Discussion The major drawback of streptokinase and urokinase is that they activate both fibrin-bound plasminogen and circulating plasminogen indiscriminately. The plasmin generated in the circulation is rapidly neutralised by the vast pool of a2-antiplasmin and is therefore lost for thrombolysis. Once the reserve of a2-antiplasmin has been exhausted, free circulating plasmin will degrade fibrinogen and other plasma proteins as well as fibrin, resulting in both a systemic fibrinolytic and a local thrombolytic state.21The danger of fibrinogen depletion is bleeding. The frequency of major haemorrhagic bleeding complications (requiring blood transfusion) in patients treated for myocardial infarction with streptokinase is less than 5% and of life-threatening haemorrhages less than 1 -5%.22 If thrombolysis is to have a therapeutic role in acute myocardial infarction, the intravenous route of drug administration is the only realistic one.23 The dose regimen of rt-PA was selected on the only therapeutic experience available when this trial was plannedland compared with a
of infusion intravenous short-term no drawback of The pretreatment having streptokinase.13 coronary angiography is that visualisation of the coronary circulation after thrombolytic treatment does not allow us to establish definite reperfusion but only to compare patency rate between 2 treatment groups. With the very strict myocardial infarction criteria applied in this trial and the short duration of clinical symptoms it is likely that 80% or more of the patients at admission had an occluded vessel in the infarct area.22,24 We chose to exclude vessels with poor distal flow from our definition of patency because clinical experience suggests that inadequate distal flow is irrelevant to
high-dose,
3
myocardial salvage.’
The 2 drug regimens were well tolerated. In 12 patients in whom the infarct-related coronary artery was still occluded after rt-PA, percutaneous transluminal coronary angioplasty (PTCA) with or without intracoronary streptokinase was carried out; 8 patients improved. Of 13 patients in whom the infarct-related artery was still occluded after systemic treatment with streptokinase, 11 had reperfusion after PTCA with or without intracoronary streptokinase. After the clinical end-point intracoronary streptokinase treatment was administered exclusively in 4 patients, of whom only 2 improved. These results are in line with an earlier study in which 6 patients unsuccessfully treated with systemic rt-PA remained completely unresponsive to intracoronary streptokinase.’1 It is not possible to analyse whether reocclusion occurred more frequently and rapidly with 1 of the 2 treatments compared in this trial since the end-point was set at the coronary angiography obtained 90 min after the start of thrombolytic treatment and further treatment was left to the discretion of the investigator. The slightly higher incidence of arrhythmia and atrioventricular block in the rt-PA-treated patients may be related to a higher reperfusion rate. 25 Various degrees of atrioventricular block are associated with reperfusion in the inferoposterior myocardium.26 The early bolus of 5000 IU of heparin together with mechanical trauma may have caused the bleeding, which occurred mainly at the arterial catheterisation site. There were appreciably fewer bleeding events in the rt-PA group. As expected, streptokinase treatment induced an intense activation of the circulating fibrinolytic system with a circulating fibrinogen level of 12% (clotting-rate assay) or 20% (precipitation method) of the starting value at the end of the infusion. With 0’ 75 mg rt-PA/kg body-weight the circulating fibrinogen at the end of the infusion fell to 61 % (clotting-rate assay) or 69% (precipitation method) of the value before infusion. A remarkable discrepancy was observed between the extent of fibrinogen degradation and the appearance of degradation products. A fall in fibrinogen of 29% at 90 min in the rt-PA group corresponded to a recovery of only 4-5% as degradation products, whereas a fall in fibrinogen of 88% in the streptokinase group was associated with 30% recovery as coagulable fibrinogen degradation products. This finding suggests that fibrinogen breakdown during rt-PA infusion is less extensive, yielding molecules which are not measured in the fibrinogen assays but are nevertheless incorporated in the clot when serum is obtained for the assay of fibrinogen
degradation products. In this trial treatment with rt-PA was associated with a higher patency rate of the infarct-related vessel; the confidence interval of the difference does not exclude the possibility that rt-PA and streptokinase are in fact of equal
847
efficacy but rt-PA treatment produced considerably systemic fibrinolysis.
less
We thank Dr W. Feuerer and the staff of the Dr Karl Thomae GmbH for
help with the preparation and dispatch of the clinical trial supplies. Correspondents should be addressed to M. V. Center for Thrombosis and Vascular Research, Campus Gasthuisberg, K.U.Leuven, Herestraat 49, B-3000 Leuven, Belgium.
BE, Geltman EM, Tiefenbrunn AJ, et al. Improvement of regional myocardial metabolism after coronary thrombolysis induced with tissue-type plasminogen activator or streptokinase. Circulation. 1984; 69: 983-90. 11. Collen D, Topol EJ, Tiefenbrunn AJ, et al. Coronary thrombolysis with recombinant human tissue-type plasminogen activator a prospective, randomized, placebocontrolled trial. Circulation 1984; 70: 1012-17. 12. Pennica D, Holmes WR, Kohr WJ, et al. Cloning and expression of human tissue-type plasminogen activator cDNA in E coli. Nature 1983; 301: 214-21. 13. Schröder R, Biamino G, von Leitner ER, et al. Systemische Thrombolyse mit Streptokinase-Kurzzeitinfusion bei akuten Myokardinfarkt. Z Kardiol 1982; 71:
10. Sobel
709-18.
Vermylen C, De Vreker RA, Verstraete M. A rapid enzymatic method for the assay of fibrinogen: the fibrin polymerization test (FPT). Clin Chim Acta 1983; 8: 418-24 Rampling MW, Gaffney PJ. The sulphite precipitation method for fibrinogen measurement; its use on small samples in the presence of fibrinogen degradation
14.
REFERENCES 1 Matsuo
15.
O, Rijken DC, Collen D. Thrombolysis by human
tissue
plasminogen
activator and urokinase in rabbits with experimental pulmonary embolus. Nature 1981; 291: 590-91. 2 Korninger C, Matsuo O, Suy R, Stassen JM, Collen D. Thrombolysis with human extrinsic (tissue-type) plasminogen activator in dogs with femoral vein thrombosis. J Clin Invest 1982, 69: 573-80. 3. Collen D, Stassen JM, Verstraete M. Thrombolysis with human extrinsic (tissue-type) plasminogen activator in rabbits with experimental jugular vein thrombosis. Effect of molecular form and dose of activator, age of the thrombus and route of administration. J Clin Invest 1983; 71: 368-76. 4. Bergmann SR, Fox KAA, Ter-Pogossian MM, Sobel BE, Collen D. Clot-selective coronary thrombolysis with tissue-type plasminogen activator. Science 1983; 220: 1181-83. 5 Van de Werf F, Bergmann SR, Fox KAA, et al. Coronary thrombolysis with intravenously administered human tissue-type plasminogen activator produced by 6 7
8
9.
recombinant DNA technology. Circulation 1984; 69: 605-10. Gold HK, Fallon JT, Yasuda T, et al. Coronary thrombolysis with recombinant human tissue-type plasminogen activator. Circulation 1984; 70: 700-07. Flameng W, Van de Werf F, Vanhaecke J, Verstraete M, Collen D. Coronary thrombolysis and infarct size reduction after intravenous infusion of recombinant tissue-type plasminogen activator in nonhuman primates. J Clin Invest 1985; 75: 84-90. Weimar W, Stibbe J, van Seyen A, Billiau P, De Somer P, Collen D. Specific lysis of an iliofemoral thrombus by administration of extrinsic (tissue-type) plasminogen activator. Lancet 1981; ii: 1018-20. Van de Werf F, Ludbrook PA, Bergmann SR, et al. Coronary thrombolysis with tissuetype plasminogen activator in patients with evolving myocardial infarction. N Engl J Med 1984, 310: 609-13.
Hypothesis OESTROGEN DEFICIENCY AFTER TUBAL LIGATION
JOHN CATTANACH 140 Power Street, Hawthorn
3122, Australia
who had undergone tubal within the past seven years were found to have oestrogen excretion concentrations at ovulation below the tenth percentile. A disturbance in the oestrogen/progesterone ratio as a consequence of localised hypertension at the ovary, when the utero-ovarian arterial loop is occluded at tubal ligation, is proposed as a possible cause of oestrogen deficiency syndrome, dysfunctional uterine bleeding, and menorrhagia after tubal ligation. Similar pathophysiology may occur after hysterectomy with ovarian conservation.
Summary
4 of 7
women
ligation
products. Clin Chim Acta 1976;
67: 43-52.
16
Merskey C, Lalezari P, Johnson AJ. A rapid simple sensitive method for measuring fibrinolytic split products in human serum. Proc Soc Exp Biol Med 1969; 131:
17.
Rijken DC, Juhan-Vague I, De Cock F, Collen D. Measurement of human tissue-type plasminogen activator by a two-site immunoradiometric assay. J Lab Clin Med 1983;
871-78.
101: 274-94. 18. Snedecor GW, Cochran WG. Statistical methods. Iowa: Iowa State
University Press,
1973: 114-16. Thomas DG, Gart JJ. A table of exact confidence limits for differences and ratios of two proportions and their odds ratios. J Am Stat Soc 1977; 72: 73-76. Gaffney PJ, Curtis AD. A collaborative study of a proposed international standard for tissue plasminogen activator (t-PA). Thromb Haemostas 1985; 53: 134-36 Collen D On the regulation and control of fibrinolysis. Edward Kowalski Memorial Lecture Thromb Haemostas 1980; 43: 77-89. Spann JF, Sherry S. Coronary thrombolysis for evolving myocardial infarction. Drugs
19.
20. 21. 22.
1984; 28: 465-83. 23. Verstraete M Intravenous administration ofa 24. 25.
26
thrombolytic agent is the only realistic therapeutic approach in evolving myocardial infarction. Eur Heart J1985 (in press). Laffel GL, Braunwald E. A new strategy for the treatment of acute myocardial infarction. N Engl J Med 1984; 311: 710-17; 770-76. Goldberg S, Greenspon AJ, Urban PL, et al. Reperfusion arrhythmia. a marker of restoration of integrated flow during intracoronary thrombolysis for acute myocardial infarction. Am Heart J 1983; 105: 26-32. Wei JY, Markis JE, Malagold M, Braunwald E. Cardiovascular reflexes stimulated by reperfusion of ischemic myocardium in acute myocardial infarction. Circulation 1983;
67: 796-801.
PATIENTS AND METHODS
Between July, 1975, and July, 1977, 601 women were instructed in the ovulation method with optional use of a diaphragm.9 They were followed up by a postal questionnaire in December, 1982. Women who had learned the ovulation method and had subsequently undergone tubal ligation were contacted. The best time to assess oestrogen status is at ovulation (Brown JB, personal
communication). To qualify for
inclusion the women had to be able to identify and the peak symptom (the last day on which cervical mucus has a distinctly lubricative quality) and be prepared to collect urine in 24-h specimen lots on the day before, the day of, and a day after identification of ovulation. Two extra women who had under-gone tubal ligation within the time limits and fitted the inclusion criteria, and had no menstrual irregularities, were also studied. Urine samples were analysed for total oestrogen excretion and associated pregnanediol in Professor E. J. Brown’s biochemistry laboratory at Melbourne University. fertile
mucus
RESULTS
13 women were eligible. 7 agreed to cooperate. In 4 of these total oestrogen levels on the day of ovulation were below the TABLE I-PUBLISHED REPORTS ON
SEQUELAE OF
TUBAL LIGATION
INTRODUCTION
No
satisfactory explanation appears to have been established for the reportedly increased incidence of menorrhagia, abnormal uterine bleeding, and major surgery after tubal ligationl-7 (table I). Since an underlying disturbance in the oestrogen/progesterone ratio is a likely explanation for these effects, I report the results of measuring concentrations of these hormones at ovulation in women who were able to use the Billings ovulation method8 to identify ovulation and who had subsequently undergone tubal ligation.
TL=tubal Available.
ligation.
Retro Retrospective.
Pros Prospective.
NA=Not