126 HYSTERECTOMY, HORMONES, AND BEHAVIOUR A Prospective Study M. BISHOP A. COPPEN Medical Research Council Neuropsychiatry Research Laboratory, We...

445KB Sizes 0 Downloads 3 Views



Prospective Study M. BISHOP


Medical Research Council Neuropsychiatry Research Laboratory, West Park Hospital, Epsom KT19 8PB, Surrey


Royal Sussex County Hospital, Brighton, Sussex G. J. R. BARNARD


Department of Obstetrics and Gynaecology, King’s College Hospital Medical School, London SE5 8RX

60 premenopausal women were assessed before and after hysterectomy for menoror fibroids or both. Their mood, sexual functioning, rhagia and plasma oestrogens and gonadotrophins were regularly assessed for a period of up to 3 years. Patients were randomly assigned to receive either oestrone sulphate or placebo tablets during the trial. No evidence was found that this group of patients showed depression or sexual difficulties related to the hysterectomy. In comparison with their baseline assessment, which was probably influenced by their gynaecological condition, they showed improved mood and vigour and unimpaired sexual activity.


Introduction DIVERSE


effects of hysterectomy have been

reported.’ Richards, in a retrospective study, found that no less than 33% of women expeiienced a depression in the 3 years following hysterectomy,2 and in a further study he reported a much higher frequency of depression, tiredness, and headaches in such patients than in women who had undergone a variety of other operations.3 The differing reports may arise from difficulty in interpreting retrospective studies which lack an adequate assessment of the patients before the operation. The present prospective study was designed to assess women before operation and to appraise them regularly for a period of up to 3 years after the operation to investigate the physical and mental effects of hysterectomy. Since psychological and sexual disturbances might be associated with altered ovarian function the investigation included assessment of plasma oestrogen, testosterone, and prolactin. Plasma-freetryptophan levels were estimated because they have been reported to be lowered in depression and to be influenced by oestrogens.4 It has also been suggested that, because of reduced oestrogen production, oestrogen replacement therapy is beneficial to post-hysterectomy patients. To clarify this issue the patients were randomly allocated to receive either oestrogen or matching placebo tablets throughout the follow-up period. Patients and Methods


60 patients who agreed to take part in the trial were randomly allocated to receive either 3 mg daily of piperazine oestrone sulphate (’Harmogen’, Abbott Laboratories Ltd.) ("active group") or identical-looking placebo tablets ("placebo group"), starting 2 weeks after their operation. All patients in the active and placebo groups were interviewed before the hysterectomy ("baseline") and at 6-week intervals during the trial. At the end of each year on the trial patients were asked to continue for another year, up to a maximum period of 3 years. 23 patients receiving oestrogen completed 1 year, 14 completed 2

years, and 7 patients 3 years. 25 patients receiving placebo completed 1 year, 14 completed 2 years, and 7 completed 3 years. 5 patients dropped out of the placebo group, including 1 patient who developed an ovarian cyst. 6 patients in the active group were not included in the analysis since their plasma oestrogen concentrations indicated that they had not taken their oestrogen tablets regularly.I patient in the active group was discontinued in the first year because of a deep-vein thrombosis. A third group of 14 patients who received neither active oestrogen nor placebo ("no treatment") was used as a control for the psychological effect of regular assessments. Apart from the usual postoperative outpatient visit this group was assessed only before the hysterectomy and 1 year after operation. The mean age of patients at hysterectomy was 40.9±0.9years in the placebo group, 42.0±1.0years in the active group, and 42.8±1.4 years in the no-treatment group. The indication for hysterectomy was either menorrhagia or fibroids of the uterus or both, except in 2 patients who had carcinoma-in-situ. All patients in the study were premenopausal at the time of their hysterectomy.


of Patients

health questionnaire (GHQ), which is used to with a non-psychotic psychiatric illness, was administered at baseline and at 6-week intervals.6 The Beck depression inventory (BDI), which is used to measure depressed mood and has been shown to correlate with clinical assessments of depression, was completed by the patient at baseline and at 6-week intervals.7 The Validity scale of the Marke Nyman temperament scale is a personality variable which refers to the energy available for psychological activity.Subjects with low Validity are cautious, tense, and easily fatigued; patients with depressive illness have low Validity even after clinical recovery. Validity was measured at baseline and after 1 year. If the patient had a partner she was asked to record the number of times she had had sexual intercourse in the previous 6 weeks (as 0,1, 4, or 8 or more times). She was also asked to rate her experience of intercourse as: (1) frequent orgasms; (2) occasional orgasms; (3) no orgasm but pleasure; (4) no pleasure; (5) active dislike of sexual intercourse. Patients were asked to record the frequency and severity of headaches and the number of sweats and hot flushes they had each day. The patient’s weight was recorded at each visit. The


identify patients

Biochemical Assessment The following hormones were estimated on a fasting venous blood sample taken between 9 A.M. and 10 A.M. at each visit. Luteinising hormone (LH), follicle-stimulating hormone (FSH) (kits obtained from CIS, U.K., Ltd.), oestrone, and 17-&bgr;-oestradiol were estimated at 6-week intervals by radioimmunoassay techniques.9 Testosterone and prolactin (CIS, U.K., Ltd.) were estimated by radioimmunoassay after 1 year. 10Free and bound tryptophan in plasma were estimated fluorimetricallylafter separation of free tryptophan by ultrafiltration 12 at 6-month intervals. The ratio of calcium to creatinine was estimated colorimetrically (kits obtained from Uniscience Ltd., Cambridge, U.K.) in a sample of morning urine collected after 1 year on the trial. The results of these investigations were expressed as the mean standard error at baseline, 6 months, 1 year, 2 years, and 3 years after hysterectomy. Group comparisons were made by paired comparisons and Student’s t-test. The mean of four estimations taken over a 6-month period was used to compare LH, FSH, and oestrogen values to minimise the variance associated with the menstrual cycle.


Psychometric Assessment No significant difference

was detected in the mean GHQ between the treatment groups at baseline, but both the active groups (after 2 years) and the placebo group (after I score


*Sigmficantly less than baseline: p<0






year) showed a significant improvement during the trial (table I). No such difference was seen in the no-treatment group. The active and placebo groups showed a significant improvement in their BDI scores after 1 and 2 years on the trial (table II). No significant change was seen in the notreatment group at 1 year. 15 patients (from both treatment groups) had baseline BDI scores greater than 10, which indicated some degree of depression. This group, whose mean BDI score was 16. O:t 1 . 1 before the operation, had a mean score of 7’5± 1-5 at 1 year (significant decrease, p<0’001). 4 patients with BDI scores greater than 10 after 1 year (mean 15 - 3±1 -8) also had scores greater than 10 before the operation (mean 18’5± 1’6). During the trial 5 patients in the active group, 4 in the placebo group, and none in the notreatment group were given antidepressant medication. The mean score on the Validity scale for patients in the two treatment groups (9 - 3:t0’ 6) was significantly lower at baseline than that for normal women (12 -2±0 -6; p<0 ’001)." After a year on the trial there was a small but significant increase from baseline (10’· 4±06; p<0 025). There was no significant difference between the treatment groups at baseline and subsequent assessment in either the frequency of sexual intercourse (table III) or frequency of orgasm reported on the questionnaire. There was no change in these variables during the trial. The numbers of headaches reported by the two treatment groups were not significantly different at any time, and the frequency of complaints of headaches did not change significantly in either group during the trial. Hot flushes were reported infrequently by patients in the two treatment groups. Hysterectomy did not make a significant difference to their frequency. At baseline 96% of the patients reported no hot flushes or less than one a day. At 1 year the comparable figure was 98%, at 2 years 83%, and at 3 years 92%. The weight of neither the active group (mean weight at baseline 61.8:tl.66 kg) nor the placebo group (mean weight 63 - 7±1 -6kg) changed significantly during the trial. Biochemical Assessment The

plasma concentrations of oestrogens at various times during the investigation are shown in table IV. The plasma oestrone concentration rose significantly in both treatment groups, and the rise was significantly greater in the active group. In contrast plasma 17-(3-oestradiol rose significantly

in both groups, but this increase was similar in the two groups. Neither LH (baseline: active group 30 ° 6±4 ·mIU/ml; placebo group 24 -7±3’7mIU/ml) nor FSH (baseline: active group 20’1±3’66 mIU/ml; placebo group 15-1±2’33 mIU/ml) differed from baseline. There was no difference between the two treatment groups at any time during the trial. At 1 year the plasma concentration of prolactin (active group 303±201 1AU/ml; placebo group 334 U/ml) and testosterone (active group 1-4±0-1 nmol/l; placebo group 1’ 3±0’1 nmol/1) did not differ between the two treatment groups. The ratio of urinary calcium to creatinine estimated after 1 year (active group 0’09±0’01mg calcium/mg creatinine; placebo group 0 -08±0 -011 mg calcium/mg creatinine) did not differ between the two groups. There was no significant difference in total and free tryptophan between the active and placebo groups at baseline or at any time during the trial (baseline total tryptophan, active group 54’4±2’0 mol/1, placebo group 54-9±1-55 mol/1; free tryptophan, active group 6-9±0-6mol/1, placebo group 6-3±0-33 J.lmol/l). There was no significant change from baseline in total or free tryptophan in the two treatment groups at any time during the trial. Discussion

investigation has shown a decrease in depression and psychiatric morbidity in patients after hysterectomy. Mood was assessed by the BDI, which correlates well with clinical ratings of mood. Both the BDI and GHQ indicated a striking improvement in psychiatric morbidity in the 3 years after hysterectomy. It is possible that the frequent assessment of patients during the investigation was therapeutic, although the no-treatment group did not show any significant change in BDI and GHQ scores after 1 year. The increase in the Validity score of patients during the trial probably indicates an improvement in their feeling of wellbeing and energy, although their mean score was significantly lower than that found in normal women. This confirms a previous report of low Validity in patients with menorrhagia. 13 No change was found in the frequency and quality of sexual intercourse after hysterectomy. The significant improvement rather than This




*Sigmficantly lower than baselme. p<0

0 I.


*Significant increase over baseline: p<0-005. Changes after I yr in patients taking active oestrogen significantly higher than in patients taking placebo for oestrone but not for (p<0-001) 17--oestroadio).


deterioration in mood is probably due to amelioration of the gynaecological symptoms which led to hysterectomy. The discrepancy between our findings and those of previous reports, which have shown an increase in psychiatric morbidity after hysterectomy, may be due to the fact that these reports were retrospective and a bias in choice of patients may have influenced these findings. A recent prospective study has shown that psychiatric morbidity is not increased by hysterectomy. 14

Endocrine studies did


hot flushes. In conclusion, we found hysterectomy to be beneficial for this group of patients. Their mood and vigour improved and their sexual life was unimpaired. The operation did not disturb oestrogen production, and we found no indication for giving oestrogen to patients in this age group. as

We thank the consultant gynaecologists in Brighton who referred their patients; the staff of the outpatients department of the New Sussex Hospital; the patients who took part in this study; Mr J. Bailey for statistical analysis; Mr H. B. Thomas, of Abbott Laboratories Ltd, for his help, and Abbott Laboratories Ltd for financial support.

Requests for reprints should be addressed to A.C. REFERENCES effects of

psychological hysterectomy. Can Psychol Rev 1977; 18: 128-41. 2 Richards DH. Depression after hysterectomy. Lancet 1973; ii: 430- 33. 3. Richards DH. A post-hysterectomy syndrome. Lancet 1974; ii. 983-85. 4. Wood K, Coppen A. The effect of oestrogens on plasma tryptophan and adrenergic function in patients treated with lithium. In: Cooke ID, ed. The role of estrogen/progestogen in the management of the menopause Lancaster MTP Press Ltd, 1978; 29-37. 5 Byrd BF, Burch JC, Vaughn WK. The impact of long-term estrogen support after hysterectomy. Ann Surg 1977; 185: 574-79. 6. Goldberg DP. The detection of psychiatric illness by questionnaire Oxford: Oxford University Press, 1972. 7. Metcalfe M, Goodman E. Validation of an inventory for measuring depression. Br J Psychiat 1965; 111: 240-42. 8. Coppen A. The Marke-Nyman temperament scale: an English translation Br J Med Psychol 1966, 39: 55-59. 9. Emment Y, Collins WP, Sommerville IF. Radioimmunoassay of oestrone and oestradiol in human plasma Acta Endocr (Kobenhavn) 1972; 3: 567-82. 10. Collins WP, Mansfield MD, Alladina NS, Sommerville IF. Radioimmunoassay of testosterone. J Steroid Biochem 1972; 3: 333-48. 11. Bloxam DL, Warren WH. Error in the determination of trypotophan by the method of Denckla and Dewey A revised procedure. Analyt Biochem 1974; 60: 621-25 12. Eccleston EG. A method for the estimation of free and total acid-soluble tryptophan using an ultrafiltration technique. Clin Chim Acta 1973, 48: 269-72. to cystic glandular hyperplasia of the 13. Bagge L A psychosomatic endometrium. Stockholm: Almqvist and Wiksell, 1963. 14. Gath DH Psychiatric aspects of hysterectomy In: Robins L, ed The social consequences of psychiatric illness New York. Brunner/Mazel, 1980 15. Janson PE, Janson I. The acute effect of hysterectomy on ovarian blood-flow. Am J Obstet Gynaecol 1977; 127: 349-52. 16. Deghenghi R. Chemistry and biochemistry of natural estrogens In: Pasetto N, Paoletti R, Ambrus JL, eds. The menopause and postmenopause. Proceedings of an international symposium held in Pome, June, 1979. Lancaster MTP Press Ltd, 1980: 3-16. 17. Schindler AE, Zwirner M, Goser R, Bolt HM Plasma concentrations of estrogens after application of various estrogen preparations. In. Pasetto N, Paoletti R, Ambrus JL, eds. The menopause and postmenopause Proceedings of an international symposium held in Rome, June, 1979. Lancaster: MTP Press Ltd, 1980: 25-30




Department of Medicine, Newcastle General Hospital, Newcastle upon Tyne NE4 6BE

reveal any evidence of ovarian

damage during the operation. This confirms earlier reports of lack of damage to ovarian blood-flow. 15 Patients not treated with oestrogen showed a small but significant increase in plasma oestrone during the trial. Plasma oestrone showed a significantly greater increase in the oestrogen-treated group than in the placebo-treated group. However, 17-0-oestradiol, which is the most biologically active oestrogen, was not significantly increased by oestrogen therapy; and this confirms previous reports.16,17 The lack of apparent biological effect of the oestrogen is probably because the patients were treatment premenopausal and showed no oestrogen deficiency. Gonadotrophins, calcium excretion, prolactin, and testosterone were within the normal premenopausal range; the patients complained of few menopausal symptoms, such

1. Meikle S The






Newcastle upon

Tyne NE1 4LP

50 diet-treated, non-insulin-dependent diabetics were tested subjectively and


objectively for chlorpropamide-alcohol flushing (CPAF) a single challenge test. Of the 12 (24%) who reported a subjective flush, 9 (18%) also flushed when a placebo was given instead of chlorpropamide, so the true incidence of chlorpropamide-alcohol flushing was 4% (1 patient was not retested with placebo). In a control group of 21 non-diabetics, 2 showed the specific CPAF phenomenon. Temperature measurement did not improve discrimination, but it did show a faster rise in facial temperature in CPAF-positive subjects than in alcohol flushers. This study does not confirm previous higher estimates of the incidence of the CPAF phenomenon in non-insulin-dependent diabetes. with

Introduction FACIAL flushing after alcohol has been estimated to occur in taking the sulphonylurea, chlorThe flush propamide.’ usually begins within 10 min of even small amounts of alcohol, reaches maximum intensity in about 20 min, and lasts about an hour.I In the single challenge test for chlorpropamide alcohol flushing (CPAF) devised by Leslie and Pyke (King’s College Hospital, London), 40 ml of sherry are given 12 h after aI single dose of 250 mg chlorpropamide.2These authors found a high degree of correlation between flushing on this test and a clinical history of alcohol flushing in patients taking chlorpropamide (33 out of 35 patients). Using this test, Leslie and Pyke found a significant association of CPAF with noninsulin-dependent diabetics (51%) when compared with insulin-dependent diabetics (10%) and non-diabetic controls 13-33% of diabetics


(10%).2 When CPAF

found to be associated with a lesser diabetic susceptibility retinopathy, ischaemic heart and vascular disease, disease,3,4 it was suggested peripheral that the single challenge CPAF test could be useful in identifying patients particularly at risk of developing these complications. We have used the same test to study the incidence of CPAF in a group of patients not treated with oral agents and drawn from the diabetic clinic of a general hospital. We have also measured changes in facial temperature during the test, in an attempt to define the temperature characteristics of the chlorpropamide-alcohol flush. was


Patients and Methods 50 consecutive patients with non-insulin-dependent diabetes (NIDDM), drawn from the diabetic clinic register of Newcastle General Hospital, were studied. Criteria for inclusion in this study were: age under 75 years; diagnosis of diabetes confirmed by oral glucose tolerance test or random blood glucose 12 mmol/t;