12 The first mechanism is unlikely because the positive association between- breast cancer and "ever" use of rauwolfia derivatives is weak and inconsistent, and there is little evidence (and none in this study) of increases in risk with increases in the amount or duration of total
VON WILLEBRAND DISEASE SAN DIEGO, A NEW VARIANT
exposure. If the last mechanism applied, breast cancers would tend to appear shortly after commencement of treatment with rauwolfia derivatives; our data do not show any evidence of this but are too few to test the possibility adequately. The data available are consistent, however, with the second mechanism-namely, promotion of the disease by rauwolfia derivatives. This should be reflected by a relative excess of rauwolfia use near to the time of diagnosis of the cancer. There is now evidence in several studies of a positive association with exposure to rauwolfia derivatives near to the time of diagnosis of breast cancer, and table vi shows evidence of increases in risk with increases in the duration of such exposure. A slight excess of "ever" use would also be expected, unless the drugs happened to be protective if given before exposure to an initiating agent. Such a slight excess was found in this study, but has not been observed consistently. It may be relevant that, in experiments with rats, treatment with reserpine before admin-
Departments of Laboratory Medicine and Internal Medicine and Clinical Investigation Center, Naval Regional Medical Center, San Diego, California 92134, U.S.A.
istration of a chemical carcinogen (dimethylbenzanthracene) reduces the incidence of mammary tumours, whereas reserpine treatment after exposure to the carcinogen enhances mammary carcinogenesis.9 It has been suggested that hypertension of long duration,10 social class, or the use of medical services4 might confound the relationship between rauwolfia derivatives and breast cancer. It was possible to examine the first two of these in this study. Hypertension of long duration showed a positive relationship with breast cancer and was also associated positively with the use of rauwolfia derivatives (table v). It is unlikely, however, that confounding on this variable could explain a specific relationship between breast cancer and the use of rauwolfia derivatives near to the time of diagnosis. High social class was also associated positively with breast cancer but not with the use of rauwolfia derivatives, and so would be unlikely to be a confounding factor. We gratefully acknowledge the assistance of Mrs D. McClelland and Mrs R. McCleese in data collection, and the cooperation of general practitioners, hospital consultants, and family-practitioner committees throughout the country. Dr H. Jick, Mr R. Peto, and Mr P. Smith gave valuable advice. B. A. was supported by a clinical sciences (epidemiology) fellowship of the National Health and Medical Research Council of Australia, and D. S. by a Rhodes scholarship. Requests for reprints should be addressed to R. D., Department of the Regius Professor of Medicine, Radcliffe Infirmary, Oxford OX2 6HE.
ADELBERT D. CRAMER
ANTHONY J. MELARAGNO
SHARLENE J. PHIFER
CECIL HOUGIE
Department of Pathology, University of California, San Diego, La Jolla,
California 92037
A reduced level of factor XII has been observed in 9 of 39 patients with von Willebrand’s disease. This apparent common variant of von Willebrand’s disease has not previously been described, and its significance is not known.
Summary
Introduction von Willebrand’s disease is an autosomally inherited haemorrhagic disease characterised by two or more of the following: prolonged bleeding-time; reduced level of procoagulant factor viu (VIII-A.H.F.) ; reduced platelet retention in glass bead columns; reduced level of factor-vm-related antigen (VIII-A.G.N.) ; reduced platelet aggregation in response to the antibiotic ristocetin (a questionable deficiency of von Willebrand’s disease); and a secondary or prolonged rise in factor VIII-A.H.F. following transfusion of haemophilic or normal plasma.1 Recent work suggests that there are several variants of the disease.1-3 Abnormal mobility of VIII-A.G.N. on twodimensional cross-immunoelectrophoresis has been demonstrated in several cases.! We report an apparent common variant of this disease in which there is a concomitant decrease in factor XII.
Patients Case1 A 65-year-old White female was admitted because of severe epistaxis. A screening partial thromboplastin time (P.T.T.) was abnormal. Fresh whole blood and posterior nasal pressurepacks were required to control bleeding. She had a history of frequent nosebleeds throughout her life. There was no family history of abnormal bleeding. Case 2 A 23-year-old Black male was admitted for circumcision. He had no history of abnormal bleeding. A presurgical P.T.T. was abnormal. 2 units of cryoprecipitate were given before circumcision. Minimal postoperative bleeding was controlled by compresses.
REFERENCES
Case3 1. Boston Collaborative
Drug Surveillance Program. Lancet, 1974, ii, 669. 2. Armstrong, B., Stevens, N., Doll, R. ibid. p. 672. 3. Heinonen, O. P., Shapiro, S., Tuominen, L., Turunen, M. I. ibid. p. 675. 4. Mack, T. M., Henderson, B. E., Gerkins, V. R., Arthur, M., Baptista, J. Pike, M. C. New Engl. J. Med. 1975, 292, 1366. 5. O’Fallon, W. M., Labarthe, D. R., Kurland, L. T. Lancet, 1975, ii, 292. 6. Laska, E. M., Siegel, C., Meisner, M., Fischer, S., Wanderling, J. ibid p. 296. 7. Armstrong, B. Int. J. Cancer, 1976, 8. Office of Population Censuses and 9. 10. 11. 12. 13.
17, 204. Surveys. Classification of Occupations.
London, 1970. Welsch, C. W., Meites, J. Experientia, 1970, 26, 1133. Magnus, K. Lancet, 1974, ii, 1080. Pike, M. C., Morrow, R. H. Br. J. prev. soc. Med. 1970, 24, 42. Miettinen, O. S. Am. J. Epidem. 1974, 100, 515 Mantel, N., Haenszel, W. J. natn. Cancer Inst. 1959, 22, 719
A 28-year-old White female was admitted for elective removal of redundant thigh fat. A screening P.T.T. was abnormal. She had undergone an ileojejunal bypass procedure 14 years previously and removal of excess panniculus 1 year previously without abnormal bleeding. No presurgical blood products were administered, and she underwent removal of thigh fat without abnormal bleeding. No abnormal bleeding was noted in the family history.
Case 4 A
58-year-old White female had
an
abnormal
P.T.T.
during
13 before open reduction of a fractured ankle. No blood products were administered, and surgery was performed without abnormal bleeding. A personal and family history revealed no abnormal bleeding. She was taking several antihypertensive medications.
COAGULATION-TEST RESULTS
tests
Case5 A 27-year-old White male presented with a history of epistaxis and easy bruisability. An abnormal P.T.T. was noted during initial coagulation testing. His mother had a history of menorrhagia, and his son had frequent epistaxis.
Case 6 A 25-year-old female presented with a history of easy bruisability. A screening P.T.T. was abnormal. The patient was lost to follow-up.
Case7 A
quet tests, and assays for coagulation factors ix, xi, and
20-year-old White
male was noted to have an abnormal P.T.T. during tests before urethroplasty and circumcision. He had a history of frequent epistaxis as a child and extreme bleeding from two lacerations. No abnormal bleeding was encountered following numerous units of cryoprecipitate before and after surgery.
xiii were
Case9 A 9-year-old Mexican male was admitted with severe epistaxis. A screening P.T.T. was abnormal. He had a 2-year history of recurrent epistaxis. The family history showed no bleeding disorders. A brother had died from acute lymphocytic leukaemia.
Methods Venous blood was anticoagulated with 3.8% sodium citrate (1 ml anticoagulant/9 ml blood). Clotting times were observed with an automatic clotting-time recorder (Medical Laboratory Automation Inc., Mount Vernon, New York). Platelet aggregation was observed with an aggregometer (Chrono-Log Corpor-
ation, Broomall, Pennsylvania). were
measured
following
a
5-min incubation period. A single lot number of commercial plasma was used for P.T.T. control (General Diagnostics, ’Verify’ normal citrate). Platelet factor in reagent plus micronised silica was also from one lot (General Diagnostics, A.P.T.T.). Factors VII-A.H.F., IX, XI, and xn were assayed by a one-stage activated P.T.T. system with congenitally deficient
plasma (Dade). The activity curve for quantitative assay was prepared from known dilutions of a single lot number of commercial plasma (General Diagnostics, ’Verify’ normal citrate). Factor VIII-A.G.N. was measured by Laurell electroimmunoassay according to the method of Zimmerman, Ratnoff, and Powell.4 Ristocetin (Abbott) aggregation of platelets was done
according to the method of Howard and Firkin.5 Platelet aggregation studies using 3 units/ml thrombin (Parke-Davis). 20 µg/ml adenosine diphosphate (A.D.P.) (Sigma), 1.9x10-5 molar connective tissue (Worthington Biochemical), and 4 g/d; adrenaline (Vitarine) as aggregating agents were performed a outlined by Weiss.6 The modified Ivy method of Mielke wa! used for determining bleeding times.’ Results
prothrombin times, thrombin times, fibrinogeI concentrations, platelet-counts, clot retractions, tourni The
A.D.P., adrenaline, and connective tissues normal except for case 4, but this patient was tested while on presurgical medications. Pertinent coagulation-test results with normal values are shown in the accompanying table. The P.T.T.S were fully corrected by diluting the patient’s plasma 1/2 with normal plasma. Measurement of factor xii by immunoassay was performed on patients 2, 6, and 8, and were 43%, 74%, and 56% respectively. Platelet retention on glass beads and patient response to transfusion were not determined.
were
A 27-year-old White male bled excessively following a tooth extraction. A P.T.T. was abnormal. A personal and family history revealed no abnormal bleeding.
P.T.T.S
platelet-aggregation studies using
thrombin,
Case8
Kieselguhr-activated
normal. The
Discussion In the patients referred to above the diagnosis of von Willebrand’s disease was based on the reduced VII-A.H.F., a reduced VII-A.G.N., and impaired platelet aggregation with ristocetin. Our results show that a significant number of patients with von Willebrand’s disease have reduced levels of factor xn. These low levels may have been previously overlooked because factor xii assays are not usually done on patients suspected of having von Willebrand’s disease. The 9 patients with reduced factor xii (table) had abnormal P.T.T.S. The 30 other patients with von Willebrand’s disease had normal factor xii and, of these, those with factor VIII-A.H.F. levels above 40% did not have abnormal P.T.T.S. Reduced factor xii could explain the observation that the P.T.T.S in von Willebrand’s disease are sometimes more prolonged than one would expect on the basis of a moderate reduction in
VII-A.H.F.
was an unreliable test for diagnosing Willebrand’s disease. This test was performed on 26 of the 39 patients in the present series and was abnormal in 6. Of these 6, all had ristocetin aggregation of platelet values greater than 25%. This disagrees with other findings where normal bleeding times were found when 3 ristocetin aggregation was greater than 25% of normal. The significance of the reduced factor xn in the pa-
Bleeding-time
von
tients of this report is not clear. Activated factor xii is believed to initiate the intrinsic coagulation, fibrinolytic, and kinin-generating pathways in human plasma.8 Factor xn is activated by collagen9 and vascular basement membrane.10 It has been proposed that normal endothelial cells synthesise and release von Willebrand factor," and synthesis of von Willebrand factor in endothelium suggests a possible relationship with activated factor xn.
14 The
relationship between VIII-A.H.F., VIII-A.G.N., and Willebrand factor is still unresolved. The presence of decreased factor xil in von Willebrand’s disease introduces yet another variable. Until further work clarifies these relationships, it is suggested that those patients who have been diagnosed as having von Willebrand’s disease and reduced factor xn be referred to as patients with von Willebrand’s disease San Diego.
von
We thank Charles G. Cochrane and Susan
Revak, Department of
Experimental Pathology, Scripps Clinic and Research Foundation, La Jolla, California, for the immunoassay measurement values. This work was supported in part by the Bureau of Medicine and Surgery Clinical Investigation Program. Requests for reprints should be addressed to A.D.C., Hematopathology Section, Department of Pathology, University of Southern Carolina School of Medicine, 2025 Zonal Avenue, Los Angeles, California 90033, U.S.A. REFERENCES
with none of the major known risk factors for the disorder. This suggests that important risk factors remain to be discovered. In a search for such risk factors, it was decided to study patients at the lower end of the age spectrum for myocardial infarction on the assumption that the evidence for a risk factor might be more exaggerated in this group and thus more easily identified. The first patient studied was a 38-year-old man who had had none of the risk factors except that he smoked. On examination there was evidence of feministion-i.e., slight gynæcomastia and rounded hips-supported by a history of loss of libido for several years and a need to shave only two or three times a week, These findings were surprising since the pronounced male prevalence of myocardial infarction in this agegroup had suggested masculinity as a factor.1 Indeed, oestrogens have been administered in an attempt to prevent
1.
I. R., Bloom, A. L., Giddings, J. C. New Engl. J. Med. 1974, 291, 113. 2. Weiss, H. J. New Engl. J. Med. 1975, 293, 580. 3. Bowie, E. J. W., Fass, D. N., Olson, J. D., Owen, C. A. Mayo Clin. Proc. 1976, 51, 35. 4. Zimmerman, T. S., Ratnoff, O. D., and Powell, A. E. J. clin. Invest. 1971,
Peake,
50, 244. 5. Howard, M. A., Firkin, B. G. Thromb. Diath. Hœmorrh. 1971, 26, 362. 6. Weiss, H. J. in Platelet aggregation (edited by W. J. Williams, E. Beutler, A. J. Ersley, and R. W. Rundler); p. 1415. New York, 1972. 7. Mielke, C. H., Kaneshiro, M. M., Weiner, J. M., Rappaport, S. I. Blood, 1969, 34, 204. 8. Kaplan, A. P. Microvaasc. Res. 1974, 8, 97. 9. Wilner, G. D., Norsl, H. L., LeRoy, E. C. J. Clin. Invest. 1968, 47, 2608. 10. Cochrane, C. G., Revak, S. D., Aiken, B. S., Wuepper, K. D., in Inflammation: Mechanisms and Control (edited by H. Lepow and P. Ward). p. 119. New York, 1972. 11. Caen, J. P., Sultan, Y. Lancet, 1975, ii, 1129.
EVIDENCE FOR HYPERŒSTROGENÆMIA AS A RISK FACTOR FOR MYOCARDIAL INFARCTION IN MEN GERALD B. PHILLIPS
Department of Medicine,
Columbia
University College
of Physicians and Surgeons, Roosevelt Hospital, New York, New York 10019, U.S.A.
Fifteen men who had had a myocardial infarction between the ages of 32 and 42 were years compared with fifteen age-matched healthy men. Seven of the patients had a strikingly slow rate of beard growth, three had evidence of gynæcomastia, and three had a loss of libido. The slow beard growth and decreased libido, and possibly the gynæcomastia, preceded the myocardial infarction. Mean serum œstradiol and œstrone concentrations were significantly increased in the patients, 43.5±8.8 (standard deviation) and 50.7±9.5, respectively, compared with 33.5±5.5 and 37.5±5.8 pg/ml in the controls (P
VON WILLEBRAND DISEASE SAN DIEGO, A NEW VARIANT
exposure. If the last mechanism applied, breast cancers would tend to appear shortly after commencement of treatment with rauwolfia derivatives; our data do not show any evidence of this but are too few to test the possibility adequately. The data available are consistent, however, with the second mechanism-namely, promotion of the disease by rauwolfia derivatives. This should be reflected by a relative excess of rauwolfia use near to the time of diagnosis of the cancer. There is now evidence in several studies of a positive association with exposure to rauwolfia derivatives near to the time of diagnosis of breast cancer, and table vi shows evidence of increases in risk with increases in the duration of such exposure. A slight excess of "ever" use would also be expected, unless the drugs happened to be protective if given before exposure to an initiating agent. Such a slight excess was found in this study, but has not been observed consistently. It may be relevant that, in experiments with rats, treatment with reserpine before admin-
Departments of Laboratory Medicine and Internal Medicine and Clinical Investigation Center, Naval Regional Medical Center, San Diego, California 92134, U.S.A.
istration of a chemical carcinogen (dimethylbenzanthracene) reduces the incidence of mammary tumours, whereas reserpine treatment after exposure to the carcinogen enhances mammary carcinogenesis.9 It has been suggested that hypertension of long duration,10 social class, or the use of medical services4 might confound the relationship between rauwolfia derivatives and breast cancer. It was possible to examine the first two of these in this study. Hypertension of long duration showed a positive relationship with breast cancer and was also associated positively with the use of rauwolfia derivatives (table v). It is unlikely, however, that confounding on this variable could explain a specific relationship between breast cancer and the use of rauwolfia derivatives near to the time of diagnosis. High social class was also associated positively with breast cancer but not with the use of rauwolfia derivatives, and so would be unlikely to be a confounding factor. We gratefully acknowledge the assistance of Mrs D. McClelland and Mrs R. McCleese in data collection, and the cooperation of general practitioners, hospital consultants, and family-practitioner committees throughout the country. Dr H. Jick, Mr R. Peto, and Mr P. Smith gave valuable advice. B. A. was supported by a clinical sciences (epidemiology) fellowship of the National Health and Medical Research Council of Australia, and D. S. by a Rhodes scholarship. Requests for reprints should be addressed to R. D., Department of the Regius Professor of Medicine, Radcliffe Infirmary, Oxford OX2 6HE.
ADELBERT D. CRAMER
ANTHONY J. MELARAGNO
SHARLENE J. PHIFER
CECIL HOUGIE
Department of Pathology, University of California, San Diego, La Jolla,
California 92037
A reduced level of factor XII has been observed in 9 of 39 patients with von Willebrand’s disease. This apparent common variant of von Willebrand’s disease has not previously been described, and its significance is not known.
Summary
Introduction von Willebrand’s disease is an autosomally inherited haemorrhagic disease characterised by two or more of the following: prolonged bleeding-time; reduced level of procoagulant factor viu (VIII-A.H.F.) ; reduced platelet retention in glass bead columns; reduced level of factor-vm-related antigen (VIII-A.G.N.) ; reduced platelet aggregation in response to the antibiotic ristocetin (a questionable deficiency of von Willebrand’s disease); and a secondary or prolonged rise in factor VIII-A.H.F. following transfusion of haemophilic or normal plasma.1 Recent work suggests that there are several variants of the disease.1-3 Abnormal mobility of VIII-A.G.N. on twodimensional cross-immunoelectrophoresis has been demonstrated in several cases.! We report an apparent common variant of this disease in which there is a concomitant decrease in factor XII.
Patients Case1 A 65-year-old White female was admitted because of severe epistaxis. A screening partial thromboplastin time (P.T.T.) was abnormal. Fresh whole blood and posterior nasal pressurepacks were required to control bleeding. She had a history of frequent nosebleeds throughout her life. There was no family history of abnormal bleeding. Case 2 A 23-year-old Black male was admitted for circumcision. He had no history of abnormal bleeding. A presurgical P.T.T. was abnormal. 2 units of cryoprecipitate were given before circumcision. Minimal postoperative bleeding was controlled by compresses.
REFERENCES
Case3 1. Boston Collaborative
Drug Surveillance Program. Lancet, 1974, ii, 669. 2. Armstrong, B., Stevens, N., Doll, R. ibid. p. 672. 3. Heinonen, O. P., Shapiro, S., Tuominen, L., Turunen, M. I. ibid. p. 675. 4. Mack, T. M., Henderson, B. E., Gerkins, V. R., Arthur, M., Baptista, J. Pike, M. C. New Engl. J. Med. 1975, 292, 1366. 5. O’Fallon, W. M., Labarthe, D. R., Kurland, L. T. Lancet, 1975, ii, 292. 6. Laska, E. M., Siegel, C., Meisner, M., Fischer, S., Wanderling, J. ibid p. 296. 7. Armstrong, B. Int. J. Cancer, 1976, 8. Office of Population Censuses and 9. 10. 11. 12. 13.
17, 204. Surveys. Classification of Occupations.
London, 1970. Welsch, C. W., Meites, J. Experientia, 1970, 26, 1133. Magnus, K. Lancet, 1974, ii, 1080. Pike, M. C., Morrow, R. H. Br. J. prev. soc. Med. 1970, 24, 42. Miettinen, O. S. Am. J. Epidem. 1974, 100, 515 Mantel, N., Haenszel, W. J. natn. Cancer Inst. 1959, 22, 719
A 28-year-old White female was admitted for elective removal of redundant thigh fat. A screening P.T.T. was abnormal. She had undergone an ileojejunal bypass procedure 14 years previously and removal of excess panniculus 1 year previously without abnormal bleeding. No presurgical blood products were administered, and she underwent removal of thigh fat without abnormal bleeding. No abnormal bleeding was noted in the family history.
Case 4 A
58-year-old White female had
an
abnormal
P.T.T.
during
13 before open reduction of a fractured ankle. No blood products were administered, and surgery was performed without abnormal bleeding. A personal and family history revealed no abnormal bleeding. She was taking several antihypertensive medications.
COAGULATION-TEST RESULTS
tests
Case5 A 27-year-old White male presented with a history of epistaxis and easy bruisability. An abnormal P.T.T. was noted during initial coagulation testing. His mother had a history of menorrhagia, and his son had frequent epistaxis.
Case 6 A 25-year-old female presented with a history of easy bruisability. A screening P.T.T. was abnormal. The patient was lost to follow-up.
Case7 A
quet tests, and assays for coagulation factors ix, xi, and
20-year-old White
male was noted to have an abnormal P.T.T. during tests before urethroplasty and circumcision. He had a history of frequent epistaxis as a child and extreme bleeding from two lacerations. No abnormal bleeding was encountered following numerous units of cryoprecipitate before and after surgery.
xiii were
Case9 A 9-year-old Mexican male was admitted with severe epistaxis. A screening P.T.T. was abnormal. He had a 2-year history of recurrent epistaxis. The family history showed no bleeding disorders. A brother had died from acute lymphocytic leukaemia.
Methods Venous blood was anticoagulated with 3.8% sodium citrate (1 ml anticoagulant/9 ml blood). Clotting times were observed with an automatic clotting-time recorder (Medical Laboratory Automation Inc., Mount Vernon, New York). Platelet aggregation was observed with an aggregometer (Chrono-Log Corpor-
ation, Broomall, Pennsylvania). were
measured
following
a
5-min incubation period. A single lot number of commercial plasma was used for P.T.T. control (General Diagnostics, ’Verify’ normal citrate). Platelet factor in reagent plus micronised silica was also from one lot (General Diagnostics, A.P.T.T.). Factors VII-A.H.F., IX, XI, and xn were assayed by a one-stage activated P.T.T. system with congenitally deficient
plasma (Dade). The activity curve for quantitative assay was prepared from known dilutions of a single lot number of commercial plasma (General Diagnostics, ’Verify’ normal citrate). Factor VIII-A.G.N. was measured by Laurell electroimmunoassay according to the method of Zimmerman, Ratnoff, and Powell.4 Ristocetin (Abbott) aggregation of platelets was done
according to the method of Howard and Firkin.5 Platelet aggregation studies using 3 units/ml thrombin (Parke-Davis). 20 µg/ml adenosine diphosphate (A.D.P.) (Sigma), 1.9x10-5 molar connective tissue (Worthington Biochemical), and 4 g/d; adrenaline (Vitarine) as aggregating agents were performed a outlined by Weiss.6 The modified Ivy method of Mielke wa! used for determining bleeding times.’ Results
prothrombin times, thrombin times, fibrinogeI concentrations, platelet-counts, clot retractions, tourni The
A.D.P., adrenaline, and connective tissues normal except for case 4, but this patient was tested while on presurgical medications. Pertinent coagulation-test results with normal values are shown in the accompanying table. The P.T.T.S were fully corrected by diluting the patient’s plasma 1/2 with normal plasma. Measurement of factor xii by immunoassay was performed on patients 2, 6, and 8, and were 43%, 74%, and 56% respectively. Platelet retention on glass beads and patient response to transfusion were not determined.
were
A 27-year-old White male bled excessively following a tooth extraction. A P.T.T. was abnormal. A personal and family history revealed no abnormal bleeding.
P.T.T.S
platelet-aggregation studies using
thrombin,
Case8
Kieselguhr-activated
normal. The
Discussion In the patients referred to above the diagnosis of von Willebrand’s disease was based on the reduced VII-A.H.F., a reduced VII-A.G.N., and impaired platelet aggregation with ristocetin. Our results show that a significant number of patients with von Willebrand’s disease have reduced levels of factor xn. These low levels may have been previously overlooked because factor xii assays are not usually done on patients suspected of having von Willebrand’s disease. The 9 patients with reduced factor xii (table) had abnormal P.T.T.S. The 30 other patients with von Willebrand’s disease had normal factor xii and, of these, those with factor VIII-A.H.F. levels above 40% did not have abnormal P.T.T.S. Reduced factor xii could explain the observation that the P.T.T.S in von Willebrand’s disease are sometimes more prolonged than one would expect on the basis of a moderate reduction in
VII-A.H.F.
was an unreliable test for diagnosing Willebrand’s disease. This test was performed on 26 of the 39 patients in the present series and was abnormal in 6. Of these 6, all had ristocetin aggregation of platelet values greater than 25%. This disagrees with other findings where normal bleeding times were found when 3 ristocetin aggregation was greater than 25% of normal. The significance of the reduced factor xn in the pa-
Bleeding-time
von
tients of this report is not clear. Activated factor xii is believed to initiate the intrinsic coagulation, fibrinolytic, and kinin-generating pathways in human plasma.8 Factor xn is activated by collagen9 and vascular basement membrane.10 It has been proposed that normal endothelial cells synthesise and release von Willebrand factor," and synthesis of von Willebrand factor in endothelium suggests a possible relationship with activated factor xn.
14 The
relationship between VIII-A.H.F., VIII-A.G.N., and Willebrand factor is still unresolved. The presence of decreased factor xil in von Willebrand’s disease introduces yet another variable. Until further work clarifies these relationships, it is suggested that those patients who have been diagnosed as having von Willebrand’s disease and reduced factor xn be referred to as patients with von Willebrand’s disease San Diego.
von
We thank Charles G. Cochrane and Susan
Revak, Department of
Experimental Pathology, Scripps Clinic and Research Foundation, La Jolla, California, for the immunoassay measurement values. This work was supported in part by the Bureau of Medicine and Surgery Clinical Investigation Program. Requests for reprints should be addressed to A.D.C., Hematopathology Section, Department of Pathology, University of Southern Carolina School of Medicine, 2025 Zonal Avenue, Los Angeles, California 90033, U.S.A. REFERENCES
with none of the major known risk factors for the disorder. This suggests that important risk factors remain to be discovered. In a search for such risk factors, it was decided to study patients at the lower end of the age spectrum for myocardial infarction on the assumption that the evidence for a risk factor might be more exaggerated in this group and thus more easily identified. The first patient studied was a 38-year-old man who had had none of the risk factors except that he smoked. On examination there was evidence of feministion-i.e., slight gynæcomastia and rounded hips-supported by a history of loss of libido for several years and a need to shave only two or three times a week, These findings were surprising since the pronounced male prevalence of myocardial infarction in this agegroup had suggested masculinity as a factor.1 Indeed, oestrogens have been administered in an attempt to prevent
1.
I. R., Bloom, A. L., Giddings, J. C. New Engl. J. Med. 1974, 291, 113. 2. Weiss, H. J. New Engl. J. Med. 1975, 293, 580. 3. Bowie, E. J. W., Fass, D. N., Olson, J. D., Owen, C. A. Mayo Clin. Proc. 1976, 51, 35. 4. Zimmerman, T. S., Ratnoff, O. D., and Powell, A. E. J. clin. Invest. 1971,
Peake,
50, 244. 5. Howard, M. A., Firkin, B. G. Thromb. Diath. Hœmorrh. 1971, 26, 362. 6. Weiss, H. J. in Platelet aggregation (edited by W. J. Williams, E. Beutler, A. J. Ersley, and R. W. Rundler); p. 1415. New York, 1972. 7. Mielke, C. H., Kaneshiro, M. M., Weiner, J. M., Rappaport, S. I. Blood, 1969, 34, 204. 8. Kaplan, A. P. Microvaasc. Res. 1974, 8, 97. 9. Wilner, G. D., Norsl, H. L., LeRoy, E. C. J. Clin. Invest. 1968, 47, 2608. 10. Cochrane, C. G., Revak, S. D., Aiken, B. S., Wuepper, K. D., in Inflammation: Mechanisms and Control (edited by H. Lepow and P. Ward). p. 119. New York, 1972. 11. Caen, J. P., Sultan, Y. Lancet, 1975, ii, 1129.
EVIDENCE FOR HYPERŒSTROGENÆMIA AS A RISK FACTOR FOR MYOCARDIAL INFARCTION IN MEN GERALD B. PHILLIPS
Department of Medicine,
Columbia
University College
of Physicians and Surgeons, Roosevelt Hospital, New York, New York 10019, U.S.A.
Fifteen men who had had a myocardial infarction between the ages of 32 and 42 were years compared with fifteen age-matched healthy men. Seven of the patients had a strikingly slow rate of beard growth, three had evidence of gynæcomastia, and three had a loss of libido. The slow beard growth and decreased libido, and possibly the gynæcomastia, preceded the myocardial infarction. Mean serum œstradiol and œstrone concentrations were significantly increased in the patients, 43.5±8.8 (standard deviation) and 50.7±9.5, respectively, compared with 33.5±5.5 and 37.5±5.8 pg/ml in the controls (P
