SERUM-ALBUMIN AND INSULIN ANTAGONISM

1192 should be told frankly that if they smoke they not put their own lives in jeopardy but, if they continue to do so during pregnancy, also expose their unborn infants to an unnecessary risk." University Department of Obstetrics and Gynæcology, Jessop Hospital for Women, C. SCOTT RUSSELL. Sheffield 3. women

only

-

COMPLICATIONS OF DEPRESSED SKULL FRACTURE I on one group of patients mentioned comment SIR,-May in the interesting paper by Mr. Miller and Professor Jennett (Nov. 9, p. 991) ? In 3 cases fractures were not elevated because the wounds had already been sutured in the casualty department. I agree that this policy of surgical inactivity is in line with traditional teaching but am fairly sure that it is high time for a change. All 3 of the Glasgow patients developed severe intracranial infection and 1 died. Although I have not analysed my own cases I can recall several in whom non-intervention once wounds have been closed has been followed by osteomyelitis, abscess formation, and meningitis. The risk nowadays of spreading infection by operating is surely far less than the risk run by accepting an unsatisfactory and dangerous situation. Naturally, one would like to think that the day may come when all compound skull fractures will be recognised and treated adequately within a few hours of wounding, but until that happy time arrives I suggest that operation should be performed on these fractures whether the skin has been sutured or not. Neuro-Surgical Unit, Middlesbrough General Hospital, Middlesbrough, Yorks.

PATRICK CLARKE.

SERUM-ALBUMIN AND INSULIN ANTAGONISM SIR,-Patients with essential diabetes mellitus have increased insulin antagonism associated with their plasma-albumin-an exaggeration of that also found in the albumin fraction of healthy people. In concentrations of 3-5-5-5 g. per 100 ml., both diabetic and normal albumin completely inhibit the effect of 1000 microunits insulin per ml. added to the incubation medium of the isolated rat diaphragm. At 1-25 g. per 100 ml., however, the albumin from essential diabetics (including prediabetics) is still highly antagonistic, whereas that prepared from healthy people is inactive.l2 Also tested in this way, the albumin from normal consanguineous relatives of diabetic patients has shown insulin antagonism.3 The albumin in the above studies was usually prepared by the trichloroacetic-acid/ ethanol method of Debro et al.,4 which is a lengthy procedure

involving dialysis for 72 hours with subsequent lyophilisation; the end-product is not completely soluble.

moreover

In 1966 Fernandez et a1.5 showed that a mixture of ethanol and hydrochloric acid gave an essentially complete precipitation of globulin from a solution of globulin and albumin. After removal of the precipitated globulin by centrifugation, the albumin was recovered from the supernatant by precipitation with sodium acetate prepared as an ethanol solution. We have adopted this principle for the preparation of albumin from the serum of normal subjects and diabetic patients. The endproduct is a finely divided off-white powder, easily soluble in water and physiological buffers, and only one protein band is obtained after electrophoresis in barbiturate buffer pH 8’6 and subsequent staining with nigrosin. Serum.-Venous blood was withdrawn into a dry syringe from fasting donors. After clotting had occurred, all samples were centrifuged for 15 minutes at 1400 g, to separate the serum. Extraction.-5 ml. volumes of serum were ordinarily used, and whenever stirring or thorough mixing was required this was effected ALBUMIN ASSAYS OF THE DIFFERENT GROUPS

ORAL CONTRACEPTIVES AND CÆRULOPLASMIN ACTIVITY SIR,-During a survey of the quantitative levels of ctrulo-

plasmin activity (assayed, using paraphenylene diamine as a substrate, in a’DU-2 ’ Beckman spectrophotometer) in the sera of individuals belonging to various races, very high values for cseruloplasmin-oxidase activity were found in women regularly receiving oral contraceptives. A comprehensive study was conducted with three control groups: (a) women of comparable age and race, married, and using other forms of contraception, such as diaphragm, intrauterine device, and safeperiod method; (b) female members of the same families as the index women; (c) three women whose levels were ascertained during and after use of oral contraceptives. The serum-casruloplasmin levels, in mg. per 100 ml. :f::S.D., was 72-75:1-7 in 12 women receiving oral contraceptives, and 34-68:3-9 in 15 women using other contraceptive methods. The female members of comparable age in the families of the index women had normal levels of serum-csruloplasmin, thus excluding the possibility of the high levels of the latter being hereditary. The serum-cxruloplasmin values in three women during and after use of oral contraceptives were as follows :

We have studied the immunochemical implications of the raised casruloplasmin level. We found that the specific activity of the protein, calculated as the ratio of mg. per 100 ml. of activity to immunochemical units of protein (estimated in immunodiffusion double-agar diffusion tubes using specific antihuman caeruloplasmin antiserum), is equal to that of normal-level cxruloplasmin. This is evidence for increased level of the protein rather than enhanced activity. Department of Human Genetics, University of Michigan, M. SHOKEIR. Ann Arbor, Michigan 48104.

*

Amount of glucose taken up incubation medium.

by diaphragm when

no

insulin is present in

by a small polytetrafluoroethylene-covered magnet. The globulin fraction was precipitated by adding the serum drop by drop to a flask containing an agitated volume of hydrochloric-acid/ethanol reagent (1 ml. concentrated hydrochloric acid in 600 ml. ethanol) equal to 9 times the volume of serum. The flask was closed with Parafilm ’ and incubated in a water-bath at 37°C for 30 minutes. The globulins were then removed by centrifugation for 15 minutes at 2600 g. Precipitation of the albumin from the isolated supernatant was effected by the addition, with stirring, of 01 M sodium-acetate/ ethanol reagent, the volume used being a fifth of the volume of the supernatant. Centrifugation at 1400 g for 5 minutes was sufficient to collect the precipitate, which was then washed with 20 ml. of methanol, followed by 20 ml. of a mixture of methanol and diethyl ether 3/1v/v. After collection by centrifugation the albumin was given a final wash with 20 ml. diethyl ether to remove the methanol. The albumin was resuspended in 10 ml. diethyl ether and collected by vacuum filtration on a sintered glass disc. When the bulk of the ether had been removed and the layer of albumin was starting to crack, the vacuum was broken and the preparation gently ground to a 1. 2. 3. 4. 5.

Vallance-Owen, J., Dennes, E., Campbell, P. N. Lancet, 1958, ii, 336. Vallance-Owen, J., Lilley, M. D. ibid. 1961, i, 804. Vallance-Owen, J. Diabetologia, 1966, 2, 248. Debro, J. R., Tarver, H., Korner, A. J. Lab. clin. Med. 1957, 50, 728. Fernandez, A., Sobel, C., Goldenberg, H. Clin. Chem. 1966, 12, 194.

1193 a fine glass rod. A further 10 ml. of ether was added, suspended, and the ether removed as far as possible by

powder using the albumin

of the vacuum. The last traces of ether were allowed evaporate by gentle grinding of the preparation. The yield of albumin was found to vary between 3-80% and 4-20%.

reapplication to

Using the rat-diaphragm-assay procedure, previously described, 67 woe have now investigated the antagonistic activity of serum-albumin prepared in this way. Three groups of adults were studied: group I, 30 overt diabetics; group II, 14 normal people who from previous observations3 89 were expected to be constituted as diabetics; and group III, a control group of 22 healthy university staff. The assay results for the different groups, shown in the

accompanying table, indicate that albumin prepared in this way is equipollent to albumin, prepared by the Debro method, regarding insulin antagonism. However, we find that the antagonism does diminish on storage and after 2 weeks has disappeared; thus the albumin should be tested within 5 days of preparation. This phenomenon, which may be due to oxidation, is being investigated. We wish to thank Dr. A. Grant and Dr. D. A. D. Montgomery for allowing us to study patients under their care and the British Diabetic Association for financial assistance. Department of Medicine, VALLANCE-OWHN

Queen’s University, Belfast.

J.

DOROTHY MCMASTER.

pital confinement for those cases who have selected as being at risk by antenatal care. With respect

been

previously

the attitude of mothers towards hospital to ask how a lay person can decide with question on which many eminent medical

to

confinement, I wish confidence persons I hope future.

on

a

disagree. to

publish

more

information

Health Department, Town Hall, Manchester M60 2JS.

on

this work in the

near

MURIEL COATES.

I comment on Sir Dugald Baird’s letter ? I have been in general practice for twenty years, attending about fifty maternity cases a year at home, but I can recall only a very small number of these patients who have had to be transferred to hospital before or during labour. This is due to understanding and confidence between patient and doctor. I can recall a large percentage of those patients (usually whose husbands insisted) who went to hospital and who later had complications. They often returned home frightened, and fearful of going through it all again. Joe Wrigley taught that the best obstetrician was the man who kept his hands in his pockets the longest time. He was right. Hospital staffs find this dictum difficult today, but it makes home confinement safe for normal cases-and most of them are. D. W. S. SHELDON. Repton, Derby.

SIR,-May

SAFETY OF DOMICILIARY MIDWIFERY

SiR,ņIshould like to comment on Dr. Galloway’s article (Oct. 26, p. 906), in which he states that there is still a place for

domiciliary midwifery, and on Sir Dugald Baird’s reply (Nov. 9, p. 1031) that the present policy of increasing the number of hospital confinements should be continued. While working with Prof. R. F. L. Logan at the Manchester Medical Care Research Unit, I was asked to review the need for hospital beds for maternity cases. The county boroughs in England and Wales were selected, since certain social data with respect to these were readily available following a previous investigation into the health of towns. The Registrar General provided information on live and still births in women of varying ages and parities and according to the legitimacy or otherwise of the children. Correlation coefficients obtained by Pearson’s ranking method revealed a highly significant degree of association between the stillbirth-rate for women of all ages and parities and the socioeconomic classification of the county borough. This applied also to housing and education. Alongside these positive associations, there was no correlation between the stillbirthrate, again for all ages and parities, and the percentage of babies delivered in hospital. When stillbirths were examined in greater detail, for mothers of parity four or more, primiparx aged 30-35 years, and illegitimate children, the stillbirthrate was not influenced by the proportion of total hospital deliveries. However, the stillbirth-rate was lower in primiparae aged 35 years and above who had hospital rather than domiciliary confinement. The same held true when all " at risk " mothers (as defined by the Cranbrook Committee) were taken together. Considering all confinements, both domiciliary and institutional, there was a highly significant relation between the stillbirth-rate and those mothers considered to be at highest risk. It is of course realised that a correlation coefficient, such as was used in this investigation, does not prove that two factors are related-it merely gives an indication. Nevertheless, this investigation did suggest that stillbirths are more closely related to social circumstances than to the proportion of babies delivered in hospital. There is still need, however, for hos6. 7. 8. 9.

Vallance-Owen, J., Hurlock, B. Lancet, 1954, i, 68. Vallance-Owen, J., Hurlock, B., Please, N. W. ibid. 1955, ii, Vallance-Owen, J., Ashton, W. L. ibid. 1963, i, 1226. Wilson, J. S. P., Vallance-Owen, J. ibid. 1966, ii, 940.

583.

ADVICE ON ABORTION SIR,-Dr. Sim’s criticism of the Pregnancy Advisory Service (Nov. 23, p. 1138) draws attention to a difficult situation. For the first time in twenty years’ general practice I am having to advise my patients to seek help outside the National Health Service for operations to which they are legally entitled. When the N.H.S. provides sufficient facilities to implement the provisions of the new Act there will be no need for organisations such as the Pregnancy Advisory Service. St. Pauls Cray, E. TUCKMAN. Kent.

BRADYARRHYTHMIA COMPLICATING MYOCARDIAL INFARCTION SIR,-The detailed descriptions of arrhythmias occurring during the first few hours after the onset of symptoms of myocardial infarction given by Dr. Pantridge and his colleagues (Nov. 23, p. 1097) are certainly a contribution to the overall appraisal of the clinical management of these patients. A comment in the discussion (para 2-" It has been suggested ...") may lead to some confusion over the views of others however, since it is not an accurate quotation and is out of context. For a representative viewpoint, and clearly for the purposes of patient management, the annotation1 should be read in full. The clinical significance of bradycardia in patients with acute myocardial infarction is of course dependent on several factors including the heart-rate, the nature and natural history of the arrhythmia, the ventricular function and strokevolume, whether bradycardia is drug-induced (e.g., morphine), and whether it responds to conservative measures such as lying the patient flat and raising the legs. Finally, I was somewhat surprised to be described as " dangerously optimistic " having been jointly responsible for an earlier paper 2 in which the clinical and haemodynamic problems of bradycardia in patients with acute myocardial infarction

were emphasised, original descriptions of the circulatory changes following atropine treatment were given,

1. Am. Heart J. 2. Thomas, M.,

1968, 75, 843. Woodgate, D. Br. Heart J. 1966, 28, 409.