Journal of Clinical Pharmacy and Therapeutics (2001) 26, 181±186 Cerebral embolism and hormone replacement therapy N. Inoue* MD PhD, K. Kihara* MD, A. Hashiguchi MD, F. Maeharaà MD PhD, S. Yoshioka§ MD and Y. Ushio MD PhD *Department of Neurosurgery, Health and Insurance, Hitoyoshi General Hospital, Kumamoto, Department of Neurosurgery, Kumamoto University Hospital, àDepartment of Radiology Health and Insurance, Hitoyoshi General Hospital, §Department of Neurosurgery, Oita Prefectural Hospital, Japan SUMMARY INTRODUCTION Few studies have focused on the relationship between hormone replacement therapy (HRT) for postmenopausal women or those with breast cancer and the occurrence of cerebral embolism. Results are con¯icting as to whether there is a link between the two. We describe three patients who experienced cerebral embolism during HRT. A 73-year-old woman had a transient ischemic attack (TIA) 6 years prior to the present admission. She then took HRT oestrogen plus medroxyprogesterone acetate for about 6 years. The HRT had been prescribed by a gynaecologist for amelioration of postmenopausal symptoms. Six years after beginning HRT, she experienced sudden onset left hemiparesis due to cerebral embolism. Two other patients had been taking HRT for breast cancers. One, a 47-year-old woman, had taken medroxyprogesterone acetate for more than one year, for recurrence of breast cancer. She had developed sudden complete left hemiparesis due to an embolism at the carotid bifurcation. The other patient, a 72-year-old woman who was taking tamoxifen citrate for prevention of breast cancer relapse, experienced cerebral embolism just 2 months after beginning tamoxifen. The risk of cerebral embolism in those on HRT should be emphasized, along with the bene®cial effects in terms of postmenopausal symptoms and prevention of breast cancer recurrence. Hormone replacement therapy (HRT), a combination of oestrogen and progesterone, is commonly prescribed to relieve symptoms of menopause and to prevent fractures in women with osteoporosis. An abundance of both clinical and experimental data suggests that HRT oestrogen in particular, has bene®cial effects on the vascular system and, at least until recently, many authorities have advocated its use for both primary and secondary prevention of cardiovascular disease. However, these recommendations have been based on observational data and no results of randomized studies looking at clinical events have as yet been published. Recently, many questions have been raised as to how oestrogen affects the cardiovascular system. Furthermore, there is confusion as to whether HRT carries more risk than bene®t (1, 2). Recent randomized trials (3, 4), which studied whether HRT can prevent coronary heart disease (CHD), indicated no ef®cacy for the overall prevention of CHD ®nding, but instead increases in the rates of thromboembolic events and gall bladder disease. By contrast, the cohort study by the Breast Cancer Detection Demonstration Project (5) concluded that oestrogen-progestin increased breast cancer risk beyond that associated with oestrogen alone in postmenopausal women. Tamoxifen is included in the category of HRT for breast cancer and is well known to be of bene®t for reducing breast cancer risk (6). Venous thromboembolic events are very common in women on HRT. However, the risk and incidence of cerebral artery infarction have not yet been clari®ed in detail. According to some reports, it appears that the relationship between HRT and the occurrence 1 of the stroke is still controversial (7±11). In this article, three cases who experienced cerebral embolism during HRT are described. Prescribing HRT requires caution, especially in terms of the Keywords: breast cancer, cerebral embolism, hormone replacement therapy, postmenopausal women Received 26 February 2001, Accepted 9 April 2001 Correspondence: N. Inoue MD, Department of Neurosurgery, Health and Insurance, Hitoyoshi General Hospital, 35 Oikami, Hitoyoshi, Kumamoto 868±8555, Japan. Tel.: +81 966 22 2191; fax: +81 966 24 2116; e-mail: ino33@hitoyoshi-hp.hitoyoshi. kumamoto.jp Ó 2001 Blackwell Science Ltd 181 182 N. Inoue et al. relationship between HRT and cerebral embolism, which has been emphasized in some reports. CASE REPORTS Case 1 A 73-year-old woman had a history of transient ischemic attack (TIA) 6 years prior to the current episode. She had left hemiparesis for about 30 min, and T2WI magnetic resonance imaging (MRI) demonstrated a high intensity area in the subcortex of the right parietal lobe (Fig. 1a). Cerebral angiography at that time showed neither a stenotic artery nor occlusion. The cardiac examination indicated no abnormalities of cardiac function. The ®ndings suggested TIA, characterized by recanalization of an embolism within a short time. Subsequently, she had been on HRT oestrogen (Premarin ˆ 0á625 mg ´ 2 per day) plus medroxyprogesterone acetate (Provera ˆ 2á5 mg per day) for about 6 years. A gynaecologist had prescribed HRT for amelioration of postmenopausal symptoms. Six years later, she experienced sudden onset left hemiparesis. Emergency cerebral angiography 90 min after the onset, con®rmed an embolism. Despite transarterial selective thrombolysis using recombinant-tissue-plasminogen acti- vator (r-tPA), the embolism could not be dissolved and spread throughout the distal portion of the proximal M1 (Fig. 1b). Three days after the onset, the right cerebral hemisphere swelled severely due to diffuse ischemia (Fig. 1c). Hence, right-sided extended external decompression was carried out in an effort to save her life. The biochemical data on admission were: prothrombin time (PT): 12á5 s, activated partial thromboplastin time (APTT): 26á2 s, INR: 1á25 and thrombo test (TT) (%): 54á6%. Protein C and Protein S activities were 106% and 57%, respectively. The patient is currently undergoing rehabilitation. HRT was discontinued. Case 2 A 47-year-old woman underwent surgery for breast cancer 5 years prior to the current episode. She was subsequently given medroxyprogesterone acetate (Hysron-H ˆ 800 mg per day) for more than one year for recurrent breast cancer. One summer afternoon, she suddenly developed complete left hemiparesis and was transferred to our department the next morning. CT scan revealed a low density area corresponding to the right MCA territory. There were no cardiac arrhythmias such as atrial ®brillation, and no thrombosis at the mitral valve Fig. 1. (a) T2WI of MRI in case 1 showing a high intensity area in the subcortex of the right parietal lobe indicating the terminal zone infarction of the right MCA territory. (b) Cerebral angiogram in case 1 showing occlusion of the right middle cerebral artery at the M1 portion. There is no ¯ow distal to this site. (c) CT scan performed 3 days after the onset in case 1 demonstrating severe swelling of the right cerebral hemisphere, with impending herniation. Ó 2001 Blackwell Science Ltd, Journal of Clinical Pharmacy and Therapeutics, 26, 181±186 Cerebral embolism and HRT except for a micro ASD con®rmed by trans-esophageal echo examination. Brain swelling was fortunately not severe in this case. External decompression was thus not necessary. A cerebral angiogram, performed 2 weeks after the onset, showed complete occlusion by a carotid artery embolism at the bifurcation in the neck without recanalization (Fig. 2a). No other part of the cerebral artery showed either stenosis or occlusion. A FLAIR image of MRI showed a high intensity area in the territory of the right MCA (Fig. 2b). The biochemical data on coagulation factors on admission were as follows, PT: 14á1 s, APTT: 39á1 s, INR: 1á08, and ®brinogen: 308 mg/dL. This patient is undergoing rehabilitation and remains on medroxyprogeterone therapy under the supervision of a surgeon. Case 3 A 72-year-old woman underwent surgery for breast cancer. She was subsequently given tamoxifen 183 citrate (Norvadex ˆ 20 mg per day) for prevention of breast cancer relapse. She experienced seizure followed by left-sided hemiparesis and left-sided space agnosia suddenly just 2 months after beginning this therapy. CT scan on admission, 4 days after the onset, showed a low density area in the subcortex of the right parietal lobe. Cardiac function examination detected no abnormalities, such as intracardiac thrombosis. A cerebral angiogram was performed 2 weeks after the onset and demonstrated only capillary blushing, suggesting the angiographical luxury perfusion as a post-phenomenon of recanalization of the embolism in the right parietal lobe (Fig. 3a). Neither stenosis nor occlusion was seen in other cerebral arteries. MRI T2WI showed a high intensity area in the parietal lobe including the subcortical area (Fig. 3b). The biochemical data on coagulation factors on admission were: PT: 11á7 s, APTT: 33á3 s, INR: 1á09, and TT: 80á4%. Protein C and S activities were slightly depressed at 53% and 51%, respectively. She Fig. 2. (a) Cerebral angiogram in case 2 performed 2 weeks after the onset showing complete occlusion of the right carotid bifurcation at the neck without recanalization. (b) FLAIR image of MRI in case 2 showing high intensity in the area of the right MCA territory. Ó 2001 Blackwell Science Ltd, Journal of Clinical Pharmacy and Therapeutics, 26, 181±186 184 N. Inoue et al. Fig. 3. (a) Cerebral angiogram in case 3 performed 2 weeks after the onset demonstrating capillary blushing, which suggests angiographical luxury perfusion as a post-phenomenon of recanalization of the embolism in the right parietal lobe. (b) T2WI of MRI in case 3 showing a high intensity area in the parietal lobe including the subcortical area. recovered almost completely, and was discharged. However, she is still taking tamoxifen for prevention of breast cancer recurrence, under the supervision of her attending surgeon. DISCUSSION We have presented three cases of cerebral embolism, in whom the embolism was probably caused by HRT despite the individual regimens being different. All had no cardiac arrythmia or stenotic lesion in both the carotid and intracerebral artery. Randomized trials (3, 4) have indicated that HRT signi®cantly increases the rate of venous thromboembolic events and that therapy with oestrogen plus progesterone has a harmful effect on blood coagulation. This is supported by other evidence (12±18). Our ®rst case had experienced a TIA 6 years prior to the current embolic event, but she was not taking HRT at the time of that initial episode. The biochemical data on coagulation factors, i.e. PT, APTT, INR, TT, and Protein C activities, were essentially within the normal ranges. Only Protein S activity was slightly low at 57%. In our second patient, who had been on medroxyprogesterone to prevent breast cancer progression, coagulation data, i.e. PT, APTT, INR, and ®brinogen levels were normal. One report (19) pointed out the discrepancy in ®ndings between clinical and biochemical data. They speculated that this discrepancy might ®rst be due to HRT inducing a procoagulant state as a result of subtle or transient changes in the coagulation system that could not reliably be detected by measuring coagulation parameters. Second, an HRT-induced procoagulant state might be attributable to an interaction between subtle changes in the coagulation system and an underlying prothrombotic disorder, for example activated protein C resistance. Our ®rst case might have had a factor, congenital or acquired, promoting thrombophilia. She had experienced a TIA before taking HRT. In additional, a micro ASD was detected by trans-oesophageal echo in our second case. Previous reports (20, 21) have pointed out the possibility of paradoxical cerebral embolism being caused by transient left-right shunt via ASD. However, there was no evidence to implicate micro ASD in this case. Changes in coagulation factors due to Ó 2001 Blackwell Science Ltd, Journal of Clinical Pharmacy and Therapeutics, 26, 181±186 Cerebral embolism and HRT medroxyprogesterone might be more likely. In our third case PT, APTT, INR and TT were within the normal ranges. Protein C and S activities were, however, slightly low at 53% and 51%, respectively. Tamoxifen, a bene®cial agent for breast cancer prevention has an action that initially decreases antithrombin III and proteins C and S during the ®rst month of treatment (22). The cerebral embolism occurred just 2 months after beginning tamoxifen. The connection between tamoxifen and stroke is not obvious, but is possible (6). In all three cases, the cerebral artery was occluded probably by an embolism caused by HRT. A relatively recent report (23) drew attention to a possible increase in the relative risk of ischemic stroke among current users of all HRT. However, nearly all review articles have suggested no signi®cant association between stroke and HRT use (7, 9±11). The mechanism of arterial occlusion by an embolism, without venous thrombosis, is not known in detail. HRT carries a greater risk of venous thrombosis and galbladder disease with no bene®cial prophylactic effect against coronary heart disease or osteoporosis for postmenopausal women (3, 4). Furthermore, HRT also increases breast cancer risk (5). In conclusion, HRT is a risk factor for cerebral embolism, especially in subjects with personal and family histories of stroke and breast cancer. The exception may be healthy relatively young postmenopausal women who have not previously experienced thromboemboli (17). REFERENCES 1. New G, Harper RW. (1999) The role of oestrogen in cardiovascular disease: bene®t or harm? MJA Practice Essentials, 171, 490±495. 2 2. Manson JE. (2000) Hormone Replacement Therapy. More risk than bene®t? Harvard Health Letter, June, 4±5. 3. Hulley S, Grady D, Bush T, et al. (1998) Randamized trial of estrogen plus progestin for secondary prevention of coronary heart disease in postmenopausal women. JAMA, 280, 605±613. 4. Grady D, Wenger NK, Herrington D, et al. (2000) Postmenopausal hormone therapy increase risk for venous thromboembolic disease. Annals of Internal Medicine, 132, 689±696. 5. Schairer C, Lubin J, Troisi R, Sturgeon S, Brinton L, Hoover R. (2000) Menopausal estrogen and estrogen- 185 progestin replacement therapy and breast cancer risk. JAMA, 283, 485±491. 6. Gail MH, Costantino JP, Bryant J, et al. (1999) Weighing the risk and bene®ts of tamoxifen treatment for preventing breast cancer. Special article. Journal of the National Cancer Institute, 91, 1829±1846. 7. Scoditti U, Buccino GP, Pini M, Pattacini C, Mancia D. (1998) Risk of acute cerebrovascular events related to low oestrogen oral contraceptive treatment. Italian Journal of Neurological Science, 19, 15±19. 8. Tavani A, Vecchia CL. (1999) The adverse effects of hormone replacement therapy. Review article. Drugs and Aging, 14, 347±357. 9. Carswell HVO, Anderson NH, Morton JJ, McClloch J, Domioniczak AF, Macrae IM. (2000) Investigation of estrogen status and increase stroke sensitivity on cerebral blood ¯ow after a focal ischemic insult. Journal of Cerebral Blood Flow Metabolism, 20, 931±936. 10. Rusa R, Alkayed N, Crain BJ, et al. (1999) 17b-estradiol reduces stroke injury in estrogen-de®cient female animals. Stroke, 30, 1665±1670. 11. Damczyk MP, Gardner DM. (2000) Risk of hormone replacement therapy in stroke Patients. Journal of 3 Clinical Pharmacy and Therapeutics, 25, 239±241. 12. McQueen EG. (1971) Hormonal steroid contraceptives. III: Adverse reactions. Review article. Drugs, 2, 20±44. 13. Buccino G, Scoditti U, Pini M, Tagliaferri SR, Manotti C, Mancia D. (1999) Low-oestrogen oral contraceptives as a major risk factor for cerebral venous and sinus thrombosis: evidence from a clinical series. Italian Journal of Neurological Science, 20, 231±235. 14. Glueck CJ, Wang P, Fontaine RN, Tracy T, SieveSmith L, Lang JE. (1999) Effect of exogenous estrogen on atherothrombotic vascular disease risk related to the presence or absence of the factor V Leiden mutation (resistance to activated protein C). American Journal of Cardiology, 84, 549±554. 15. Levi M, Middeldorp S, Buller HR. (1999) Oral contraceptives and hormonal replacement therapy cause an imbalance in coagulation and ®brinolysis which may explain the increased risk of venous thromboembolism. Cardiovascular Research, 41, 21±24. 16. Whiteman MK, Cui Y, Flaws JA, Espeland M, Bush TL. (1999) Low ®brinogen level: a predisposing factor for venous thromboembolic events with hormone replacement therapy. American Journal of Hematology, 61, 271±273. 17. Koh KK, Horne MK, III, Cannon RO III. (1999) Effects of hormone replacement therapy on coagulation, ®brinolysis, and thrombosis risk in postmenopausal women. Thrombosis and Haemostasis, 82, 626±633. 18. Baumann P, Diedrich K. (2000) Thromboembolic complications associated with reproductive endo- Ó 2001 Blackwell Science Ltd, Journal of Clinical Pharmacy and Therapeutics, 26, 181±186 186 N. Inoue et al. crinologic procedures. Hematology/Oncology/Clinics of North America, 14, 431±443. 19. Douketis JD, Gordon M, Johnston M, Julian JA, Adachi JR, Ginsberg JS. (2000) The effects of hormone replacement therapy on thrombin generation, ®brinolysis inhibition, and resistance to activated protein C: prospective cohort study and review of the literature. Thrombosis Research, 99, 22±34. 20. Cheng TO. (1976) A diagnostic challenge and its detection during life. Circulation, 53, 565±578. 21. Jones RJ, Caplan LR, Come PC. (1983) Cerebral emboli of paradoxical origin. Annals of Neurology, 13, 314±319. 22. Oberhoff C, Szymeczek J, Hoffmann O, Winkler UH, Kaiser S, Schindler AE. (1998) Adjuvant antiestrogen treatment with tamoxifen in postmenopausal women with breast cancer: a longitudinal study of blood coagulation and ®brinolysis. Breast Cancer Research and Treatment, 50, 73±81. 23. Grodstein FG, Stampeer MJ, Manson JE, et al. (1996) Postmenopausal estrogen and progestin use and the risk of cardiovascular disease. New England Journal of Medicine, 335, 453±461. Ó 2001 Blackwell Science Ltd, Journal of Clinical Pharmacy and Therapeutics, 26, 181±186