ON SOME LATE EFFECTS OF BILATERAL OOPHORECTOMY IN THE AGE RANGE 15-30 YEARS
B. W. Johansson, L. Kaij, S. Kullander, H .- C. Lennér, L. Svanberg and B. Åstedt
From the Department of Gynaecology (Head: S. Kullander, M.D.), the Heart Laboratory of the Department of Medicine (Head: B. Hood, M.D.) and the Department of Psychiatry (Head: C. G. Dahlgren, M.D.) at the University of Lund, Allmänna Sjukhuset, Malmö, Sweden
Abstract. Between 1910 and 1940, 146 young females, aged 15-30 years, underwent bilateral salpingo-oophorectomy as part of a radical operation for salpingo-oophoritis. These women or their records were reviewed in 1971. 42 women had died in the meantime. More than half (22) of them had died from cardiovascular diseases, 5 from carcinoma of the uterus and 4 had committed suicide. None had died from carcinoma of the breast. Of 68 who were still alive, informa- tion by questionnaire was obtained and 32 were admitted to hospital for extensive examination. 32 age-matched women to be operated on for prolapse but with no other known disease of the reproductive tract served as controls. A further control group was added as 11 of the 68 were found to have menstruated again after the operation which had evidently not completely removed the gonads. Complete oophorectomy was found to have been followed by: (a) an increased incidence of cardiac symptoms and nervous dis- eases as well as an increased use of drugs; (b) a significant increase in the frequency of coronary vascular diseases in ages up to 70 years; (c) an increase in the serum cholesterol and triglycerides, most significantly in the ages below 60-65 years. Women with symptomatic coronary disease had a higher serum cholesterol level than women without and women with signs of peripheral vascular diseases had a significantly higher concentration of serum triglycerides: (d) an increased frequency of fractures (radius and femoral neck), increased osteoporosis and thinner cortical bone. The brittleness of the skeleton was correlated with low excretion of oestrogens in the urine. No vertebral compres- sion or abnormal decrease in height was observed. (e) an increased adrenocortical activity with significantly in- creased excretion of 17-ketosteroids, 17-OH-ketosteroids and low polar total oestrogens. This activity abated in women above 65 years. (f) a traumatic psychological ex- perience of the accompanying sterility while sexuality seemed to be largely unaffected in many of them. Almost half of the women examined by the psychiatrist were unusually mentally active and agile and they had a lower excretion of estriol than the rest.
The effect of the ovaries on the panorama of diseases and causes of death of women has long attracted
attention. Most published investigations have been based on women oophorectomised for a variety of reasons.
It is believed that bilateral oophorectomy has a favourable effect on several types of tumour, mainly by the elimination of most of the oestrogenic hormones. Such hormones are thought to play a role in the genesis and further growth of tumours of the reproductive tract and of the breast. Oophorectomy has been tried also in the treatment of carcinoma of the breast.
Bilateral oophorectomy with a substantial reduc- tion of oestrogens is stated to be followed by an increased frequency of disease of the cardiovascular system with an increased tendency to arterial hypertension and arteriosclerosis and an ac- companying tendency to coronary sclerosis (64, 53), myocardial infarction (41) and cerebral haemorrhage (4, 49). Symptoms of coronary disease are more common in women after oophorectomy (53). Coro- nary atheroma is more common in men than in women up to the age of 60 years, after which the difference narrows. This difference in incidence, like the increase in the frequency of symptomatic coronary disease (CHD) after the menopause, cor- roborates the assumption that the higher blood level of oestrogens in women with preserved gonadal function offers protection against the development of coronary atheroma. It has also recently been shown that the menopause is more often premature in women who have had myocardial infarction than in the population in general (7).
The composition of lipids in the serum has re- ceived much space in the discussion of the develop- ment of arteriosclerosis over the last 20 years. The
serum cholesterol level is much higher in women than in men during the sixth decade of life. Oo- phorectomy prior to menopause raises the serum cholesterol level (3, 42, 43, 53, 58). Oestrogenic hormones evidently have a strong influence on the lipid and lipoprotein content of the blood (59). Oestrogens increase the phospholipid concentra- tion, suppress cholesterol, raise the alpha- lipoprotein level and depress the beta-lipoprotein. In women who have been oophorectomised or who are in the climacteric, the lipid and lipoprotein pattern resembles that in men. But treatment of such women with oestrogenic hormones will cause the levels of the above serum fractions to return to normal.
Electrocardiograma changes are abnormally common after oophorectomy (43), but the frequency of such changes can be reduced by treatment with oestrogens (15). This lends further support to the idea that oestrogens protect women against cardiosclerosis.
Various endocrine organs are affected by loss of ovarian function. Such loss has been assumed to cause a generally increased hypophyseal activity with an increase not only of the production of gonadotropins but also of other hormones, such as ACTH and TSH. It has been claimed that the adrenal cortex can increase its activity in oopho- rectomised women and help to maintain the pro- duction of oestrogens. It is possible that oopho- rectomy may be followed by disturbances of thyroid function, by a tendency to diabetes and an increased frequency of obesity.
The classical symptoms of loss of the ovaries (i.e. symptoms due to the loss of oestrogens) are various neurovegetative symptoms, such as flushes. Mental symptoms such as lack of initiative, apathy, and depression or increased activity, restlessness or unrest may also occur. In addition, anxiety and psycho-sexual aberrations have been reported (46).
The almost complete loss of the production of sex hormones after oophorectomy seems to have an effect also on the cortex and trabeculae of the bones. This results in osteoporosis with an increased risk of fracture after even relatively trivial trauma. But osteoporosis is an insidious condition which is not easy to diagnose early. The condition is not radiolog- ically demonstrable until the skeleton has lost at least 30% of its calcium content, i.e. not before the process is well advanced. As early as 1941 Albright et al. (1) postulated that the loss of endocrine func- tion of the ovaries was a contributory cause of
osteoporosis in women, but this assertion was later denied by Donaldson & Nassim in 1954 (17) who stressed the importance of ageing. Increased para- thyroid activity due to loss of oestrogenic suppres- sion might well explain the increased serum calcium and increased excretion of calcium in the urine as well as the negative calcium balance (65). But also increased serum phosphorus levels have been re- ported (1) which would rather suggest decreased parathyroid function.
Oestrogens are excreted, though in low concentra- tion, in the urine also after complete oophorectomy. Such oestrogens, as a rule <10 µg a day (16) are believed to originate mainly from the adrenals, since administration of ACTH to oophorectomised women increases the excretion of oestrogens in the urine (11).
After the menopause the excretion of oestrogens in the urine is normally about 10 to 15 µg per day, which means a rate of production that is no longer capable of stimulating the endometrium (12). A fall in the level of the oestrogens in the climacteric or after oophorectomy results in an increase in the blood and urine level of gonadotropins, mainly of FSH but also of LH (44). The excretion of 17-ketosteroids, 17- ketogenic steroids, pregnanetriol, testosterone and corticoids decreases in old age-the adrenopause -but does not appear to be correlated with ovarian function.
The urogenital manifestations of loss of produc- tion of oestrogens include atrophy of the mucosa of the vagina, urethra and bladder with decreased re- sistance to infections. The uterus as well as any existing myoma becomes smaller and the en- dometrium becomes thin and inactive. The cervix becomes less protuberant and the fornices are shal- low.
In many departments of gynaecology, hyste- rectomy, carried out for benign diseases of the reproductive tract, is often extended to include routine oophorectomy, especially in women of pre- menopausal or menopausal age in an endeavour to prevent the later development of ovarian tumours. But such an operation, however, appears to be capa- ble of leading to climacteric symptoms even in women after the menopause if the oestrogenic production by the ovaries exceeds 20 u per 24 hours (39).
In earlier published series of oophorectomised women the majority were operated on late in the reproductive period of life (4).
Many women in the district covered by Malmö general hospital have been oophorectomised early in life, however. This is because early in the century advanced salpingitis was often treated surgically at our hospital. Not only the tubes but also the ovaries were radically removed. This surgical treatment which was performed even on girls as young as 15 years of age, was abandoned with the advent of sulphonamides at the end of the 1930s.
This unique series was analysed in respect of morbidity, survival, and causes of death. Those patients who were still alive and residing in the Malmö district were admitted to the hospital and investigated with regard to nervous disorders, cardiovascular status, osteoporosis and hormonal state.
MATERIAL AND METHODS
The 1910-40 files of Malmö General Hospital were searched for the names of women operated on because of salpingo-oophoritis with salpingectomy and bilateral oo- phorectomy before the age of 30 years. With the help of the parish offices and registry offices most of the women still living could be traced. The entire series consisted of 146 women who had operations between the ages of 15 and 30 years (mean age 25 years). Of these women, 42 had died in the meantime. The causes of death were noted either from the death certificates or sometimes from the autopsy reports. Of the 104 survivals, who at the time of the present review in 1971 were, on average, 69 years old 68 were living in the area served by Malmö General Hospital and were contacted. 32 (group A) of these women, aged 53-84 years who had their opera- tions at an average age of 24 years, and who were living within the town were invited to a free gynaeco- logical and physical examination. All were accepted and all were admitted to the hospital. The remaining 36 (group B) were sent a questionnaire, which they were asked to fill in and return. All cooperated. As controls we used 32 randomly selected, age-matched women (group C) admitted to the department for surgical treatment of pro- lapse of the uterus, but with no other disease of the repro- ductive organs.
Of the 68 living women who had been operated and who were still alive, 4 of group A and 7 of group B continued to menstruate after the operation, which had thus not re- moved all ovarian tissue. This group of incompletely treated women-group D (DA+DB)-thus constituted a second control group of 11 cases. In the tables the figures given for groups A and B do not include women belonging to group D.
The questions in the questionnaires (which were also posed to the patients admitted to hospital) concerned civil status, adopted children, later treatment at hospital (and if so, when, where and why), bone fractures, later medical examinations for heart symptoms, arterial hypertension, vessel cramp, nervous symptoms or any other disease.
The women admitted to hospital were examined routine- ly, besides which determinations were made of the total
oestrogens according to Carlström et al. (14), 17- ketosteroids, with the method of Vestergaard (62) and Birke et al. (8) and 17-OH-ketosteroids in the way de- scribed by James et al. (26) in 24-hour samples of the urine. Urine voided under sterile conditions was cultured for bacteria. At the gynaecological examination vaginal smears were obtained and examined for malignant cells as well as for the effect of oestrogens.
The thickness of the cortical bone of the proximal shaft of the radius was examined radiologically according to Meema (36) to assess the extent, if any, of osteoporosis. The serum calcium and phosphorous were also deter- mined.
Out of the 32 women of group A, 20 were randomly selected and thoroughly examined psychiatrically on the basis of a 90 minute traditional psychiatric interview, de- signed to cover relevant aspects of subjects’ lives, with special emphasis on psychiatric and psychosexual factors. It would have been impossible to evaluate the controls blindly, and as the estimated variables were, to a large extent, subjective, it was decided not to include the con- trols in the psychiatric part of the investigation.
The women were examined cardiologically for coronary disease and stenosing arterial disease of the legs. Determi- nations were also made of the serum lipids. In the analysis of the material two age limits were used, one at 75 years and one at 65 years.
A special search was always made for angina pectoris and intermittent claudication with the criteria published by WHO. A 12-lead electrocardiogram (ECG) was recorded and in most patients an exercise test was performed with the patient in the sitting position on an electrically braked variable-load bicycle ergometer (Elema-Schönander, Stockholm-Solna). The initial work load was 200 kpm per min. and if a steady state was achieved within 4 minutes the load was increased to 400 kpm per min and thereafter, if possible, to 600 kpm per min. The ECG was recorded on a direct ink writing machine (Mingograph 81; Elema- Schönander).
The following leads were recorded with the patient supine before exercise: I, II, III, aVR, aVL, aVF, V1, V2, V3, V4, V, and V7.
During exercise the reference electrode was placed on the forehead and leads CH2, CH4, CH, and CH7 were recorded. CH-leads were also recorded before and im- mediately after exercise, while the patient was sitting on the bicycle, and 5 and 10 minutes, respectively, after exercise, while the patient was resting supine. Two min- utes after exercise with the patient resting supine, a complete 12-lead electrocardiogram, including V-leads, was made. The heart rate during exercise was, as a rule, calculated every minute from the ECG. The systolic blood pressure was measured every other minute, and just before the end of exercise, by the auscultatory method with sphygmomanometer cuff wrapped around the right arm. The respiration rate was measured with a stethoscope every other minute. In the analysis of the ECG only ST depressions of 1 mm or more (measured from the end of the P-R segment) and horizontal or downward sloping ST segments were considered abnormal.1 1 Working ECG was judged by two cardiologists independ- ently of one another.
| Age (year) at death | Cancer | Cardio- vascular disease | Tubercu- losis | Violent death | Miscella- neous | Number |
|---|---|---|---|---|---|---|
| 25-30 | 1 | 1 | ||||
| 31-35 | 1 | 1 | 2 | |||
| 36-40 | 1 | 1 | 2 | |||
| 41-45 | 2 | 1 | 3 | |||
| 46-50 | 1 | 3 | 4 | |||
| 51-55 | 3 | 3 | 6 | |||
| 56-60 | 2 | 2 | ||||
| 61-65 | 1 | 5 | 1 | 7 | ||
| 66-70 | 1 | 1 | 2 | |||
| 71-75 | 1 | 4 | 1 | 6 | ||
| 76-80 | 5 | 1 | 6 | |||
| 81-85 | 1 | 1 | ||||
| . Total | 9 (22 %) | 22 (52 %) | 3 (7%) | 6 (14 %) | 2 (5%) | 42 |
| Mean age at death | 54 | 65 | 33 | 55 | 65 |
Cardiac decompensation was classified according to the criteria of the New York Heart Association. Heart volume, with the patient standing, was measured ac- cording to Jonsell (27) and relative heart volume, i.e. ml per square meter body surface area (ml/m2 BSA) was estimated from a radiograph according to Lys- holm et al. (33).
Serum triglycerides was determined by the method of Laurell (31) and the total cholesterol was measured in a Technicon Auto Analyzer (10, 30, 45).
Hypotheses were tested with Student’s t-test and calcu- lation of x2, sometimes with Yate’s correction (57).
RESULTS
Causes of death (42 cases)
The causes of death given in the death certificates and autopsy reports of the 42 women who died
| Number of patients | Age (year) Mean age | |
|---|---|---|
| Cancer | ||
| Bronchial carcinoma | 1 | 71 |
| Pharyngeal carcinoma | 1 | 64 |
| Carcinoma of the pancreas | 1 | 57 |
| Carcinoma of the cervix | 2 | 54 |
| Carcinoma of the uterine | 3 | 44 |
| body | ||
| Cardiovascular diseases | ||
| VOC+cor incomp angina | 3 | 56 |
| Myocardial infarct | 3 | 57 |
| Cardiosclerosis | 4 | 69 |
| Cerebro-vascular accident | 5 | 68 |
| Arteriosclerosis | 3 | 66 |
| Arteriosclerosis+ Parkin- sonism | 1 | 74 |
| Pulmonary embolism | 3 | 75 |
before our review are distributed among the main groups in Table I. A list of those who died from carcinoma and from cardiovascular diseases is given in Table II.
In 22 % of the 42 patients who died, the cause of death was carcinoma. The corresponding figure for the whole country in 1968 was roughly the same or 19.5% (60). It is noteworthy, however, that in 5 of the 9 cases the tumour was in the uterus, which seems to be an over-representation of this organ, whereas there was not a single case of carcinoma of the breast. Early oophorectomy thus appears to offer protection against carcinoma of the breast.
The frequency of deaths from cardiovascular dis- eases, especially cerebrovascular lesions, myocar- dial infarction and cardiosclerosis, was high, sug- gesting that oophorectomy early in life accelerates development of atheroma.
The frequency of death from tuberculosis, how- ever, was 7%, compared with 0.7% for the country as a whole in 1968. This difference can probably be explained by the marked decrease in the frequency of tuberculosis during recent years and by the fact that some of the patients operated on because of salpingitis had tuberculous salpingitis.
The frequency of violent deaths was 14% (6 cases), compared with 6.5% for the whole country in 1968. Of the 6 patients, as many as 4 had committed suicide (oophorectomy at 21, 23, 29 and 30 years). This means that almost 10% of the women without ovaries who died com- mitted suicide, which is a remarkably high figure. The cumulative age- and sex specific
| Group A+B | Group C | Group D | |
|---|---|---|---|
| Number of patients | 57 | 32 | 11 |
| Married | 47 | 30 | 10 |
| Children before opera- | |||
| tion | 22 | 29 | 2 |
| Adopted children | 10 | 0 | 3 |
| Group A+B (57) | Group C+D (43) P | ||
|---|---|---|---|
| Biliary disorders | 20 | 11 | n.s. |
| Cardiac disorders | 23 | 8 | sign .* |
| Hypertension | 20 | 16 | n.s. |
| Intermittent claudica- | |||
| tion | 6 | 4 | n.s. |
| Nervous complaints | 22 | 7 | sign .* |
| Fractures | 23 | 10 | n.s. |
| Various other disorders | 32 | 22 | n.s. |
| Receiving medication | 38 | 17 | sign .* |
risk of suicide during 1920-69, based upon the national incidence (9), was calculated as 0.64 for the entire sample. The observed number, 4, was thus more than 6 times that expected. However, it should be kept in mind that many of the patients had many social problems at time for operation. As early as 1962 Ask-Upmark (4) reported 2 cases of suicide among 38 deaths in a review of women who had both ovaries removed. These women had, however, been operated on relatively late in life (at 40 and 45 years).
Survivors examined
Anamnestic data
It was striking that many of the women had married after their operations and were evidently mentally and sexually adjusted to married life despite sterility, absence of ovaries and menstruation (Table III). Many of them had adopted children.
The control group in whom ovarian ablation had been incomplete (D) was so small that in further comparisons it was pooled with the group of controls
| Oophorec- tomised (n=41) | Controls (n=29) | |
|---|---|---|
| Myocardial infarction | 3 | 0 |
| Angina pectoris | 8 | 3 |
| Pos. working ECG, not AP | 0 | 3 |
| Total CHD | 11 | 6 |
| Cor incomp. (no CHD, no VOC) | 4 | 0 |
| Periph. arteriosclerosis | 5 | 3 |
| Hypertension | 5 | |
| (diast. BP ≥100 mmHg) | 8 |
| Oophorec- tomised (n=17) | Controls (prolapses) (n=17) | |
|---|---|---|
| Infarction | 1 | 0 |
| Angina pectoris | 4 | 0 |
| Pos working ECG, no AP | 0 | 2 |
| Total CHD | 5 | 2 |
| Periph. arteriosclerosis | 3 | 0 |
with prolapse (C). It was found that biliary symp- toms (gallstone and/or inflammation of the gall blad- der) was not much more common among women who had complete removal of the ovaries while cardiac symptoms and nervous complaints were significantly more common in this group. Complete oophorectomy probably also leads to an increased frequency of fractures (almost significant), various other diseases and the need for medication.
Special cardiovascular findings, Blood lipids
The frequencies of various cardiovascular findings are given in Table IV.
It is clear from the Table IV that no certain differ- ence in frequency was found between the groups. In patients below 65 years, however, the frequency of Coronary Heart Disease (CHD) and peripheral symptomatic arteriosclerosis tended to be higher in the oophorectomised group, but the number of pa- tients was small and the difference was not statisti- cally significant (Table V).
The group of patients with signs of coronary dis- ease included those who had no subjective symp- toms suggesting CHD but in whom the working ECG suggested coronary insufficiency. It is noteworthy that if those patients are not included in the CHD- -group which thus then consisted only of patients with frank myocardial infarction and angina pectoris, the difference in the frequency of coronary disease in ages below 70 years was significant (Table VI).
No difference in blood pressure was found with certainty between the women after oophorectomy and the controls (Fig. 1).
No change in the serum cholesterol or serum triglycerides could be demonstrated with certainty after oophorectomy in the series as a whole (Table VII).
In the lower ages, i.e. below 60-65 years, how-
| Age | Oophorec- tomised | Con- trols | Diff. | |
|---|---|---|---|---|
| ≤60 | MI+AP | 2 | 0 | p<0.05 |
| Miscellaneous | 3 | 9 | ||
| ≤65 | MI+AP | 5 | 0 | p<0.05 |
| Miscellaneous | 12 | 17 | ||
| ≤70 | MI+AP | 9 | 1 | p<0.05 |
| Miscellaneous | 20 | 22 | ||
| ≤75 | MI+AP | 11 | 3 | No sig- |
| Miscellaneous | 30 | 26 | nificance | |
ever, the cholesterol and triglyceride levels were higher in the women after oophorectomy than in the controls. The difference was significant for serum cholesterol in ages below 60 and for serum triglycerides in the group below 65 years (Figs. 2 and 3).
The serum lipid level in the oophorectomised pa- tients with CHD did not differ from that in the controls with CHD. Neither was any significant difference found in serum lipids between controls with and without demonstrable coronary disease. But oophorectomised patients with CHD had a higher serum cholesterol level than those without CHD, 283 and 249 mg/100 ml, respectively (p<0.05).
When the postoperative group and the control group were pooled, the cholesterol level was higher in patients with signs of CHD than in those without, but no notable difference was found in serum triglyceride concentration (Table VIII).
In contrast with what was found in the pooled material, the serum triglyceride levels in patients with signs of peripheral vascular disease was higher than in patients without, while no clear difference was found in the serum cholesterol level (Table IX).
Osteoporosis
The records showed an almost significant increase in the frequency of fractures in the women who had been castrated. Detailed analysis of the case histories revealed that the frequency of fractures of
| Oophorec- tomised (n=41) | Controls (n=29) | P | |
|---|---|---|---|
| Cholesterol, mg/100 ml | 260 | 251 | n.s. |
| Triglycerides, mmol | 1.24 | 1.09 | n.s. |
BLOOD PRESSURE
mm Hg 200
OOPHORECTOMISED
CONTROLS
180
160
140
120
100
80
0
51-55 56-60 61-65 66-70 71-75
YEARS OF AGE
the radius was significantly increased (Table X). Alffram & Bauer (2) found a markedly increased frequency of radial fractures after the menopause in the women in Malmö. In our group of women after oophorectomy this increase is pronounced. Oophorectomy may thus accelerate osteoporosis, a common predisposing cause of fractures; 2 of the women had had 3 radial fractures and 1 woman had had 2. The 15 radial fractures reported had oc- curred in 10 women. Fractures of the femoral neck showed the same tendency to increase in frequency as radial fractures, while the frequency of vertebral compression and fractures showed no such tend- ency. Thus body height did ot differ between post- operative women-160 cm (150-171) and the con- trols-157 (146-168). Byt osteoporosis was signifi- cantly more pronounced in the oophorectomised women.
Oophorectomy had thus increased the tendency to fractures, osteoporosis and weakness of the bones.
The cortex, as measured by the method of Meema (36), was found to be thinner in the postoperative group (Table XI). The serum phosphorous, but not the serum calcium, was for some unknown reasons increased in the women after oophorec- tomy.
Among the women after oophorectomy, the skele- ton was less brittle (fewer fractures) when the excre- tion of oestrogens was higher and that of ketosteroids lower and this was found both in women above and below 65 years.
CHOLESTEROL
mg %
320
OOPHORECTOMISED
300
☒ CONTROLS
280
260
240
220
0
51-55 56-60 61-65 66-70 71-75
YEARS OF AGE
Hormonal balance
Table XIII compares the excretion of hormones in the postoperative women and in the controls.
The significantly increased activity of the adrenals after oophorectomy is noteworthy. It was most pro- nounced in the younger group of women. Also the oestrogens excreted had probably been derived from the adrenals. It may therefore be a sign of adaptation and of the defence mechanism of the body initiated by the gonadectomy.
Asymptomatic bacteriuria was found in 2 of the 32 sterilised women examined. Cytological examina- tion of smears revealed no signs of malignancy in 29 of the 32 women, but atypia in 3 of them. None of the smears showed signs of increased oestrogenic influ- ence (acidophilia).
Psychiatric findings
The psychiatric examination was limited to 20 of the 32 thoroughly investigated patients (Group A), summarised in Table XIV. Only 3 (15%) were single and 3 were divorced. None of these figures differ from those for the general population. Seven women
| CHD (n=17) | No CHD (n=53) | |
|---|---|---|
| Cholesterol | 280 mg/100 ml | 249 mg/100 ml p≤0.05 |
| Triglycerides | 1.26 mmol | 1.15 mmol n.s. |
| P.A. (n=8) | No P.A. (n=62) | |
|---|---|---|
| Cholesterol | 266 mg/100 ml | 255 mg/100 ml n.s. |
| Triglycerides | 1.50 mmol | 1.13 mmol p≤0.05 |
were mothers at the time of the operation; 4 of them had their children out of wedlock.
In retrospect, the most important result of the operation in the women’s opinion, was the sterility. Twelve (60%) women said that the thought of the sterility was a terrible shock, while 8 said it caused no problems or even a relief. In the sample as a whole there was no relationship between childlessness and regret of sterility. Within the group of women who had children before their operation, however, the 3 women with legitimate children, regretted the sterili- ty, while the 4 with illegitimate children did not. This latter distribution is hardly due to chance (exact p=0.028).
Judging from the brief review, the psychological mechanisms used by the women for coping with the trauma of sterility varied. Some of those who did not regret the condition evidently used denial and re- pression of their sorrow, anger and grief. Two women indicated that they accepted the sterility as a “punishment” for their illegitimate preg- nancies.
Two women reported phobic reactions (fear of
TRIGLYCERIDES m mol
2.00-
1.80
OOPHORECTOMISED
☒ CONTROLS
1.60
1.40
1.20
1.00
0
51-55 56-60 61-65 66-70 71-75
YEARS OF AGE
| Group | ||
|---|---|---|
| A+B | C+D | |
| Total number of patients in group | 57 | 43 |
| Patients with fractures | 23 | 10 |
| Number and type of fractures | ||
| Radius | 15 *** | 2 |
| Femoral neck | 8 | 3 |
| Vertebral column | 2 | 1 |
| Others of known site | 5 | 1 |
| Others of unknown site | 2 | 5 |
| Total number of patients exam. in regard of osteoporosis | 28 | 43 |
| Number of patients with Rtg. osteoporosis | 6 *** | 1 |
being alone and fear of going out, respectively) after their operation. One had fainting attacks for 10 years after the operation. Epilepsy was suspected, but not proved, and several years later she reacted to domestic stress with clear-cut hysterical reaction. She could not stand or walk. One woman used projection: she blamed her former fiancé in a very unrealistic way for having caused her illness. None of the women blamed the doctors or the hospital. To one woman the sterility was the tragedy of her life. She was not informed about it until 3 years after the operation-it was a great shock, she became de- pressed, she gave up her fiancé and her career and moved to a small place until pension age. At the investigation she was still very unhappy. At least 6 of the women clearly described how they coped with the crisis, they found a satis- factory substitute in work, education, creative activities or adopted children.
At the examination 7 women presented symptoms and/or signs of nervous disorders. The diagnoses were as follows.
| Group | Corticalis- thickness (mm) | Ca/s (mmol/l) | P/s (mg/100 ml) |
|---|---|---|---|
| A | 4.3 (n=13) | 4.71±0.18 (n=16) | 3.55 *** ±0.62 (n=17) |
| C | 4.6 (n=16) | 4.80±0.24 (n=26) | 3.05 ±0.63 (n=24) |
| Pat. No | Age | Diagnosis |
|---|---|---|
| 1 | 64 | Anxiety neurosis |
| 3 | 68 | Subnormality |
| 4 | 56 | Depression |
| 5 | 69 | Subnormality +personality disorder |
| 7 | 59 | Character neurosis |
| 10 | 71 | Depression |
| 12 | 75 | Cerebral arteriosclerosis |
Patients 1 and 4 had had their nervous symptoms since the operation, No. 10 since she had been informed that she was infertile. For the remaining 4 subjects there was no time relationship between the operation and the symptoms.
The personalities and complaints of 5 women ap- peared normal for age, 8 (40 %) women were unusu- ally vital, active and sympathetic.
The sexual activity of the women could be as- sessed in all cases except one, in which the patient refused to discuss the matter (Table XV). One women had had no sexual relationship after the birth of her child, 3 years before the operation. Three women lost their libido after the opera- tion. The husband of one of them died the same year as her operation and she still mourned him. In her second marriage she was frigid. It is not possible to decide whether the operation or the bereavement was responsible for her frigidity. Two more women became frigid later in life:
| Group A | Oestrone (µg/24 hours) | Oestradiol (mg/24 hours) | 17-keto- steroids (mg/24 hours) | 17-OH-keto- steroids (mg/24 hours) |
|---|---|---|---|---|
| Castrated patients with fractures | ||||
| (n=5) <65 years | 2.7 | 6.5 | 7.9 | 12.0 |
| (n=7) >65 years | 2.0 | 4.4 | 5.3 | 8.4 |
| Castrated patients without fractures | ||||
| (n=10) <65 years | 3.0 | 6.1 | 6.6 | 10.8 |
| (n=6) >65 years | 3.1 | 9.6 | 3.7 | 8.1 |
| Group | Low polar total oestrogens (µg/24 hours) | 17-keto-steroids (mg/24 hours) | 17-OH-keto-steroids (mg/24 hours) |
|---|---|---|---|
| Castrated (A) (n=15) <65 years | 11.2 | 7.01 p *** p *** 4.6J | ! 11.2) *** *** 8.2 p |
| (n=13) >65 years | p *** 8.9 | ||
| Controls (C) | *** | p | |
| (n=17) <65 years | P 3.5 | 4.4 | 8.2 |
| (n=15) >65 years | 2.0 | 4.6 | 10.8 |
one when she was informed of the infertility, one after a hysterectomy 19 years after castra- tion. The sexual activity of the majority of the women thus seemed largely unaffected by the operation.
With but one exception there was no relationship between the psychiatric variables and the results of the hormone analyses. The exception was that in the women who appeared to be unusually mentally ac- tive the excretion of oestriol was lower (diff. 1.867, t=2.508,p<0.25).
DISCUSSION
Earlier comparisons of the incidence of CHD be- tween oophorectomised women after oophorectomy and controls have given differing results. In an au- topsy series consisting of 49 women who had bilat- eral oophorectomy 2-42 years previously. Wuest et al. (64) found more coronary atheroma than in their control series. But their series were small, and the range of variation of the severity of the coronary changes was wide. The difference in frequency between the series and the controls was largest in the subgroup that had undergone
| Age | |
| mean | 65 years |
| range | 54-85 years |
| Civil status | |
| unmarried | 3 15% |
| married before op. | 6 |
| married, total | 17 |
| divorced after op. | 3 15% |
| Women with children | 7 |
| thereof illegitimate | 4 20% |
| Admits sterility being a trauma | 12 60% |
oophorectomy 5-9 years before death. In another post-mortem investigation, Rivini and Dimitroff (52) reported a significantly higher frequency of pronounced atheroma of the coronary vessels in oophorectomised women than in the gen- eral population. In a clinical investigation of such women compared with a personally selected control group consisting of women after hysterectomy as well as with an age-corrected group from the popula- tion of Framingham, Robinson et al. (19) found a significantly higher frequency of arteriosclerosis, manifested clinically as CHD or peripheral vascular disease. In a similar investigation Oliver & Boyd (43) found a higher concentration of cholesterol and coronary vascular disease more often after bilateral than after unilateral oophorectomy.
In contrast with the above-mentioned in- vestigators, however, Novak & Williams (41) found no difference post mortem in the frequency of marked arteriosclerosis between women after oophorectomy and controls. But their series in- cluded women who had had their operations only a few years before death as well as women who had reached such an age that the difference, if any, between the two groups must have been at least partly masked by senile arteriosclerosis. If the women of advanced age be excluded, the frequency of severe atheromatosis in the postopera-
| Before operation | After operation | |||
|---|---|---|---|---|
| Good | Poor | No exp. ? | ||
| Good, excellent | 10 | 3 | - | - |
| Poor | - | 2 | - - | |
| No experience | 3 | - | - - | |
| Unknown | - | - | 1 1 | |
tive group will be higher than in the controls. In a well controlled investigation, Ritterband et al. (51) did not find the frequency of arterio- sclerotic heart disease to be increased in women after oophorectomy. But they reported a re- markably low frequency of CHD, about 9%, in their group compared with usually about 20% in clinical studies by other investigators.
The marked increase in the frequency of coronary vascular disease in women after 60 years of age occurs about 15 years after the natural menopause. This increase in CHD is ascribed to the decrease in the synthesis of oestrogens at the onset of the menopause. It was therefore considered legitimate to study the incidence of CHD in women whose ovaries had been removed at least 15 years previous- ly. On the other hand, the increased occurrence of arteriosclerotic vascular changes in advanced age will probably mask any difference between normal and oophorectomised women regarding clinical signs of vascular disease of the heart and limbs if the interval is too long. In our investigation this interval was, on average, 43 years. This long interval proba- bly masked the difference suspected in the total group in the frequency of CHD and in the concentra- tion of the blood lipids apparent in the lower age groups.
The difference observed in the serum cholesterol level between women with and without signs of CHD is well known. Our finding that stenosing peripheral arteriosclerosis correlates better with the high triglyceride level than with the high cholesterol level is supported by earlier investigations (22, 25, 28, 56). Neither in the Framingham investigation (19) of pa- tients above 50 years was the cholesterol concentra- tion found to be higher in patients with intermittent claudication than in those without.
The investigations referred to above and observa- tions made in the present study appear to warrant the conclusion that bilateral oophorectomy leads to a higher frequency of myocardial infarction and an- gina pectoris. Like Robinson et al. (53), we found that if the criteria of the clinical diagnosis CHD are extended to include patients with a doubtful history or patients without symptoms but with specific ir- regularities in electrocardiograms recorded during work, it will diminish the difference between women with and without ovaries. Asymptomatic persons with a working ECG suggesting coronary insuffi- ciency, however, develop manifest coronary vascu- lar disease more often than persons with a normal
working ECG (5, 35, 54). These observations might suggest that oophorectomy accelerates coronary stenosis and symptomatic coronary disease. In fact, in our series many of the younger women with such a disease in the postoperative group had already died. This means that the group of survivors was selected in respect of CHD.
The most comprehensive psychiatric review of castrated women that has hitherto been published is that by Pedersen (46). The present study differs from Pedersen’s in two respects: our series included only women whose operations were before the age of 30 and the interval between the operation and the re- view was almost 4 times as long (means 41.5 years, compared with 11.4 years).
During the interval between operation and follow-up, almost one-third of the women in the original sample died. From a psychiatric point of view this selection affects women with cerebral arteriosclerosis and those who committed suicide. The effect of this bias on the result is uncertain. The main subjective effects of the operation are menopause, the hot flushes and the sterility. In this review sterility was by far the most important effect and the one which to a large extent determined the total effect of the operation on the lives of the women. To some women, probably those whose lives had been less fortunate and whose social condi- tions had been less satisfactory, sterility meant a catastrophe or life-long neurotic adaptation. To other women it was a severe psycho-traumatic crisis, but one which they could cope with and they eventually found gratification in substitute activities. In a broad psychological sense of the term oophorectomy does not differ from other kinds of psycho-traumatic crises.
Pedersen refutes Bleuler’s assumption of a general “Endocrines Psychosyndrome” in women after oophorectomy, and proposes the term “climacteric psychosyndrome” for the vasomotor disturbances with hot flushes, sweating, tachycardia etc., which are common features of both the normal and the artificial climacteric and due to oestrogen deficien- cy.
In the present sample the immediate mental reac- tions to the operation were ascribed to the trauma of sterility. This is high-lighted by a woman who was normal until 3 years after the operation, but when she was informed of the sterility she became depressed.
Our impression from this small sample examined
is, then, that the nervous disorders following oophorectomy are of psychological, rather than of endocrinological, origin.
It was noteworthy that judging from the histories and the personal interviews, many of the women were unusually active, energetic, and sthenic. This is in agreement with Pedersen’s study, where 41 % of the total sample was described as “restless”, “agile” and “hyperactive”. The most probable explanation of this observation is a psychologi- cal one, i.e. women deprived of the possibility of giving birth direct their libidinal energy (in the psychodynamic sense of the word) to the outer reality. On the other hand, the finding in the present study that the group of women judged to be unusually active had a lower excretion of oestriol indicated that the high level of mental activity of these women was associated with the endocrinological effect of the castration. This assumption is supported in Pedersen’s study, in which the sthenic traits were significantly more often observed in totally than in subtotally castrated women (49 and 15%, resp. x2=15.63, 1 d.f.).
Pedersen’s explanation that the sthenic traits were secondary to ANS (autonomic nervous system) symptom (hot flushes etc.) does not hold true for the present sample as most subjects had passed their period of hot flushes and no association could be found between duration of severity of these symp- toms and the level of mental activity. Three alter- native hypotheses may, perhaps, explain the effect of castration on the mental activity.
Ovarian gestagens (progesterone) are known to have a sedative effect on the CNS, at least in large doses (37). It has been described as the “passivity” hormone, preparing the women to the self-centred state of pregnancy (6). A life-long deprivation of progesterone might allow for a higher level of mental activity.
Low oestrogen level induces high levels of FSH which has a stimulatory influence on the CNS of the rabbit (55).
Castration increases the level of FSH-RF (gonadotropin releasing factor) in hypothalamus of rats (34, 38). Whether this neurohormone has any influence on higher brain centres or on mental func- tioning is not known, but recently a stimulating and mood-elevating effect of TSH-RF (thyrotropin re- leasing factor) has been demonstrated, both in nor- mal (48, 63) and in subjects with endogenous depres- sion (29, 47). If FSH-RF has a similar effect or, if
castration has an effect on the synthesis and release of TSH-RF the resulting increased mental activity might be explained through the hypothalamus.
The absence of a gross effect of castration on sexual function finally confirms earlier observa- tions. ’
The increased “vicarious” adrenal activity in these women was noteworthy. It has often been shown that the adrenal cortex produces steroid sex hormones and undergoes proliferative changes in gonadectomised animals (for ref. see Thung (61)). Frantz and Kirschbaum (20) who studied gonadectomised mice, observed that in some strains the pattern of the adrenocortical hor- mones was dominated by oestrogen; in others, by androgen.
The adrenal cortex and its reaction to stress is of significance in coronary diseases. It may also help to explain the osteoporosis and weakness of the bones. As known, rarefaction of bone occurs and gradually progresses in all women above 40 years of age. This senile osteoporosis is thus more marked after the menopause. Extensive decalcification with pa- thological osteoporosis cannot be explained by ageing alone. It is of endocrine origin probably with involvement of the sex hormones. Such hormones are produced not only by the ovaries but also by the adrenal cortex. Primary or ac- companying functional disorders of the adrenals may thus contribute to such osteoporosis. Cal- citonin might secondarily be involved in this mechanism (24).
The “protective” mechanism which women after oophorectomy have been deprived of is the monthly secretion of oestrogens-gestagens as well as the profound changes of pregnancy (prolonged hae- modynamic adjustments etc.). This loss in such women, on the other hand, means elimination of factors probably playing a significant role in the cau- sation of mammary tumours. Remarkably enough, none of these women had died from mammary carcinoma and of those alive only one of them had been operated on for such a condition. Earlier ex- perience has shown that women who have oophorectomy before the age of 40 develop cancer of the breast only one-fourth as often as women in comparable groups (19, 23).
Hormonal influence is no doubt operative in the initial phase of development of breast cancer. En- docrine abnormalities-low androgen metabolite excretion-may be a primary factor long before
breast cancer develops (13). The extent to which prolactin is of importance in the pathogenesis and growth of breast cancer is at the moment a matter of debate. Prolactin production probably is severely reduced during the life of women castrated when young. Its blood level is now being studied in the group of women described and so are other gonadotropins and steroids. Deep-frozen blood and urine specimens are available for continued re- search.
In the light of our observations and earlier know- ledge it would seem wise, as suggested, to substitute oestrogen in the artificial as well as in the natural menopause (21). Synthetic oestrogens may· imply a certain risk of thrombosis. Natural oestro- gens seem to be preferable (66), especially since they also have a suppressive effect on triglyceri- des and ß-lipoproteins (32). Atrophic changes in the urogenital tract are counteracted by oestro- gens, which also have a beneficial, well docu- mented effect, on ageing skin (50). Prophylactic treatment with oestrogen can prevent the develop- ment of oesteoporosis (40) and should be useful in the prevention of psychic disorders owing to lack of oestrogen.
It has been proposed that after the menopause and throughout the rest of life most women should re- ceive substitution therapy with natural oestrogens and progesterone added to induce regular monthly bleeding (21). The control of the effect of such a treatment could then very well be included in regular periodic gynaecological health checks at which vag- inal smears should be examined not only for malignant cells but also for the effect of oestrogen. Such follow-ups should, of course, also include de- termination of the serum cholesterol and triglyceride levels, hepatic enzymes, E.C.G. and determination of the bone density. At these regular check-ups close attention should be given to any occurrence of abnormal growths in the mammary glands, too.
REFERENCES
1. Albright, F., Smith, P. H. & Richardson, A. M .: JAMA 116: 2456, 1941.
2. Alffram, P. A. & Bauer, G. C .: J Bone J Surg 44A: 150 1962.
3. Aitken, J. M., Lorimer, A. R., Mckay Hart, D., Lawrie, T. D. V. & Smith, D. A .: Clin Sci 41: 597, 1971.
4. Ask-Upmark, E .: Acta Med Scand 172: 129, 1962.
5. Brody, A. J .: JAMA 171: 1195, 1959.
6. Benedek, T. & Rubinstein, B. D .: Psychosom Med 1: 245, 1939.
7. Bengtsson, C .: Acta Med Scand, suppl. 549, 1973.
8. Birke, G., Diczfalusy, E. & Plantin, L. O .: J Clin endocrinol Metab 18: 736, 1958.
9. Bolander, A. M. in Suicide and Attempted Suicide. Scandia Intern. Symposia, Stockholm, 1972.
10. Boyd, J., Bonnafé, M. & Mazett, J. B .: Ann Biol Clin 10: 12, 1960.
11. Brown, J. B., Falconer, C. W. A. & Strong, J. A .: J Endocrinol 19: 52, 1959.
12. Brown, J. B. & Matthew, G. D .: Recent Prog Horm Res 18: 337, 1962.
13. Bulbrook, R. D .. Hayward, J. L. & Spicer, C. C .: Lancet II: 395, 1971.
14. Carlström, K. & Furuhjelm, M .: Acta Obstet Gynecol Scand 50: 259, 1971.
15. Davis, M. E., Jones, R. J. & Jarolim, C .: Am J Obstet Gynecol 82: 1003, 1961.
16. Diczfalusy, E. & Lauritzen, C .: Oestrogene beim Menschen. Springer Verlag, Berlin, 1961.
17. Donaldson, J. A. & Nassim, J. R .: Br Med J I: 1228, 1954.
18. Feinleib, M .: J Nat Cancer Inst 41: 315, 1968.
19. Framingham study. Robinson, R. W., Higano, W., Cohen, W. D .: Arch Int Med 104: 908, 1959.
20. Frantz, M. J. & Kirschbaum, A .: Cancer Res 9: 257, 1949.
21. Furuhjelm, M .: Sv Läkartidn 71: 605, 1974.
22. Greenhalgh, R. M .: Lancet II: 947, 1971.
23. Hirayama, T. & Wynder, E. L .: Cancer 15: 28, 1962.
24. Hollo, I .: Therapia Hung 4: 123, 1972.
25. Isacsson, S. O .: Thesis. Lund 1972.
26. James, V. H. T. & Caie, E .: J Clin Endocrinol Metab 24: 180, 1964.
27. Jonsell, S .: Acta Radiol (Stockh) 20: 325, 1939.
28. Kannel, W. B., Dawber, Th. R., Skinnel, J. J. Jr, MacNamara, P. M. & Shurtleff, D .: Circulation 32: 4, suppl. II, II-121, 1965.
29. Kastin, A. J., Schalch, D. S., Ehresing, R. H. & Anderson, M. S .: Lancet II: 740, 1972.
30. Kessler, G .: Adv Clin Chem 10: 56, 1967.
31. Laurell, S .: Scand J Clin Lab Invest 18: 668, 1966.
32. Lebech, P. E. & Borggaard, B .: In Klimakteriet. Fre- deriksberg 1973.
33. Lysholm, E., Nylin, G. & Quarna, K .: Acta Radiol (Stockh) 15: 237, 1934.
34. Martini, L., Frashini, F. & Motta, M .: Recent Progr Horm Res 24: 439, 1968.
35. Mattingly, T. W .: Amer J Cardiol 9: 395, 1962.
36. Meema, H. E .: J Canad Ass Radiol 13: 27, 1962.
37. Merryman, W .: J Clin Endocrinol 14: 1567, 1954.
38. Mittler, J. C. & Meites, J .: Endocrinology 78: 500, 1966.
39. Mölsted-Pedersen, L., Lebech, P. E. & Sikjär, B .: In Klimakteriet, Frederiksberg 1973.
40. Nordin, B. E. C .: Br Med J I: 571, 1971.
41. Novak, E. R. & Williams, T. J .: Am J Obstet Gynecol 80: 863, 1960.
42. Oliver, M. F. & Boyd, G. S .: Lancet II: 1273, 1956.
43. Oliver, M. F. & Boyd, G. S .: Lancet II: 690, 1959.
44. Papanicolaou, A. D., Loraine, J. A., Dove, G. A. & London, N. B .: J Obstet Gynecol Br Comm 76: 308, 1969.
45. Pearson, S., Stern, S. & McGavak, Th. II .: Anatyl Chem 25: 813, 1953.
46. Pedersen, A. L .: Kirurgisk klimakterium (Surgical climacterium), Århus 1956.
47. Prange, A. J. Jr, Lara, P. P., Wilson, I. C., Alltop, L. B. & Breese, G. R .: Lancet II: 999, 1972.
48. Prange, A. J. Jr, Wilson, I. C., Lara, P. P., Wilber, J. S., Breese, G. B., Alltop, L. B. & Lipton, M. A .: Arch Gen Psych 29: 28, 1973.
49. Randall, C. L., Paloncek, F. P., Graham, J. B. & Graham, S .: Am J Obstet Gynecol 88: 830, 1964.
50. Rauramo, L. & Punnonen, R .: Arch Gynäkol 201: 202, 1971.
51. Ritterband, A. B., Jaffe, I. A., Densen, P. M., Magag- na, J. F. & Reed, E .: Circulation 27: 237, 1963.
52. Rivini, A. U. & Dimitroff, S. P .: Circulation 9: 533, 1954.
53. Robinson, R. W., Higano, N. & Cohen, W. D .: Arch Int Med 104: 908, 1959.
54. Rumball, A. & Acheson, E. D .: Br Med J 1: 423, 1963.
55. Sawyer, C. H. & Kawakami, M .: Endocrinology 65: 622, 1959.
56. Slack, J .: Lancet II: 1380, 1969.
57. Snedecor, G. W .: Statistical Methods. Ames, Iowa, 1948.
58. Sperry, W. M. & Webb, M .: J Biol Chem 187: 107, 1950.
59. Svanborg, A. & Vikrot, O .: Acta Med Scand 179: 615, 1966.
60. Sveriges Officiella Statistik: Dödsorsaker 1968. Statistiska Centralbyrån, Stockholm, 1970.
61. Thung, P. J .: Gerontologia 6: 41, 1962.
62. Vestergaard, P .: Acta Endocrinol (Kbh) 8: 192, 1951.
63. Wilson, I. C., Prange, A. J. Jr, Lara, P. P., Alltop, L. B., Stikelaether, R. A. & Lipton, M. A .: Arch Gen Psych 29: 15, 1973.
64. Wuest, J. H., Dry, T. J. & Edwards, J. E .: Circulation 7: 801, 1953.
65. Young, M. M. & Nordin, B. E. C .: Lancet II: 118, 1967.
66. Åstedt, B. & Jeppsson, S .: J Obst Gynaec Br Comm 81: 723, 1974.
Submitted for publication July 15, 1974
S. Kullander Department of Gynaecology Allmänna sjukhuset S-214 01 Malmö Sweden