EARLY DETECTION OF RELAPSE IN HYPERTHYROIDISM BY T.R.H. TEST

661 THYROTROPIN, THYROXINE, AND TRIIODOTHYRONINE

LEVELS IN TWO INFANTS WITH CONGENITAL HYPOTHYROIDISM

9 after 1 min. and 10 after 5 min. The child had a diffuse goitre, approximately 5 cm in diameter. With a rapid semiquantitative method for serum-thyrotropin analysis, high values were found in the cord-serum. Daily sampling for the analysis of serum thyrotropin, thyroxine, and triiodothyronine was then performed until replacement

was

initiated. X-ray examination revealed two ossification than 1 mm in size, in the right distal femoral epiphysis. Electroencephalography revealed normal response to a stimulus by light but increased latency-time-an effect of thyroid insufficiency, as indicated by normalisation of the time during replacement

therapy

was

centres, not more

against estimation of thyroxine 899 in screening for congenital hypothyroidism and in favour of thyrotropin estimation. University Departments of Clinical Chemistry and of Pædiatrics, Sahlgren’s Hospital, S-413 45 Gothenburg; and Department of Pædiatrics, G. LINDSTEDT LIN Central Hospital, K. IVE IVERSEN S-431 22 Mölndal, Sweden. M. OSKARSSON OSK

therapy (Dr T. Hrbeck). Apart from minor feeding difficulties and a somewhat slow weight there were no symptoms or signs of hypothyroidism. Normal variations in bilirubin concentration were noted. Serum-thyrotropin levels were high, whereas the serum-thyroxine concentration, initially somewhat low, decreased slowly in the ensuing days (see accompanying table). The concentration of serum-triiodothyronine was normal in cord-blood: the post-natal rise was abnormally low. Treatment with triiodothyronine was therefore started after 6 days.

gain,

patient, a boy, is the second child of a 42-year-old who had previously had primary sterility for 7 years. The first child, also a boy, appeared normal at 6 months of age and had a normal value for his length. Growth retardation was noted during the following year, length—1.S.D. at 12 months and -2 S.D. at 20 months of age. A diagnosis of hypothyroidism was then established. At that time, his mother had been bearing the second child for 3 months. Because of suspected hypothyroidism she received thyroxine therapy CASE 2-This

woman

from that time.

Delivery was by ctsarean section in the 40th week of gestation. Apgar score was 8 after 1 min and 9 after 5 min. Birth-weight was 3590 g, length 50 cm. The infant had no goitre, nor any other symptom or sign of thyroid disease. A high concentration of serum thyrotropin was noted after 4 days (see table). X-ray examination of distal femur at day 6 revealed an ossification centre of 3 mm size. Thyroxine therapy was started at day 8. The child also had Down’s syndrome with trisomy 21 and a congenital cardiac disorder-possibly atrial septal defect.

In summary, it appears that both children had reduced

capacity in the thyroid. Possibly, the formation of thyroid hormones had been nearly adequate during fetal life as witnessed by the presence of ossification centres in the distal femur as well as only moderately decreased values for serum-thyroxine concentrations in the cord-blood (case 1) and at day 4 (case 2). Apart from the goitre in one of the patients, there was no clinical evidence of a thyroid disorder. However, E.E.G examination in one child revealed cerebral dysfunction. A normal rise of serum thyroxine and triiodothyronine concentrations could not be demonstrated in case 1, nor was there any evidence for a normal rise in case 2. reserve

Thus, in these two cases X-ray examination of bone

centres

the distal femur failed to disclose thyroid insufficiency. Analysis of serum thyroxine and triiodothyronine gave definitely pathological values during the lst week in only one child, whereas the thyrotropin concentration was abnormal in

EARLY DETECTION OF RELAPSE IN HYPERTHYROIDISM BY T.R.H. TEST

SIR,-Ormston et al. 10 reported long-term follow-up studies

patients with Graves’ disease after discontinuation of antithyroid drug therapy. They reported several hormonal pat- , terns of relapse with premonitory or permanent and isolated or simultaneous increases in T3 and or T4’ They concluded that serial estimations of thyroid hormones, mainly T,, provide the most reliable method of monitoring relapse in hyperthyroidism.

of

Serial estimations of thyroid hormones must be done over a to distinguish between the different hormonal patterns of rebound and relapse. Isolated increases of either hormone can only be detected by measuring both T, and T4. This procedure is laborious, time-consuming, and expensive, and will delay the diagnosis of relapse.

prolonged period

We believe that the thyrotrophin-releasing hormone (T.R.H.) is the best and most simple procedure for early detection of recurrence in hyperthyroidism. This test is the most sensitive method of detecting all forms of hyperthyroidism, even at the preclinical stage.1O-11 A T.R.H. test must be performed before stopping antithyroid treatment; after a period of several months (18 months as a rule), a clear T.s.H. response is seen in almost all euthyroid patients (some euthyroid patients do not respond to T.R.H,15 16 but this is mainly observed at the beginning of antithyroid treatment when T.s.H. reserve is not yet restored). When hyperthyroidism recurs, the T.s.H. response is suppressed very early independently of the hormonal pattern and T, and T 4’ test

experience, if T3 or T rises because of the rebound phenomenon, the T.s.H. response to T.R.H. is not suppressed but is generally augmented. By this simple test, recurrence of hyperthyroidism can be detected even several months before clinical symptoms develop. If T.S.H. is suppressed, the relapse In

our

in

both cases. We do not know if laboratory investigation in the neonata period by existing methods may disclose late-appearing thyroic insufficiency--e.g., due to dysgenesis or enzyme defect. It is o interest that case 2, in whom analysis of thyroxine and triio dothyronine gave near-normal results, had a brother who de veloped hypothyroidism at the age of a year. This might speal

8. Dussualt,

J. H., Coulombe, P., Laberge, C., Letarte, J., Guyda, H., Khoury, J. Pediatrics, Springfield, 1975, 86, 670. 9. Larsen, P. R., Broskin, K. Pædiat.Res. 1975, 9, 604. 10. Ormston, B. J., Garry, R., Cryer, R. J., Besser, G. M., Hall, R. Lancet, 1971, ii, 10. 11. Haigler, E. D., Jr., Pittman, J. A., Jr., Hershman, J. M., Baugh, C M. J. clin. Endocr. Metab. 1971, 33, 573. 12. Besser, G. M., Mortimer, C. H. J. clin. Path. 1974, 27, 173. 13. Shenkman, L., Mitsuma, T., Hollander, C. S. J. clin. Invest. 1973, 52, 205. 14. Staub, J. J. Lancet, 1975, i, 868. 15. Clifton-Bligh, P., Silverstein, G. E., Burke, G. J. clin. Endocr. Metab. K.

1974, 38, 531. 16. Sanchez-Franco, F., Garcia, M. D., Cacicedo, L. ibid. p.1098.

662 can

be

proved by single T and and

treatment

or

T

measurements

before

is started.

SIR,—We have been interested for

Endocrine Unit, Medizinische Universitätspoliklinik, and Department of Nuclear Medicine, Kantonsspital, CH-4004 Basle, Switzerland

J. J. STAUB P. L. BARTHE I. WERNER

Department of Endocrinology, University Children’s Hospital, CH-4005 Basle,

J. GIRARD

Switzerland

ULTRASOUND AND MAMMALIAN D.N.A.

SIR,—Because of increasing medical use of ultrasound, especially in obstetric diagnosis, it is important to check that it is not genetically harmful. Most studies of the effect of sonication at medical doses on chromosomes have shown no damage. Nevertheless, the absence of visible chromosomal damage does not exclude the possibility of damage to the D.N.A. molecules. We therefore checked solutions of purified calf-thymus D.N.A. for possible effects of ultrasound treatment applied within a range of frequencies and intensities used in medicine, both in treatment and in obstetric diagnosis. The D.N.A. molecules were photographed in an electron microscope (magnification x 29 000), and the lengths of sonicated and control molecules of D.N.A. ’

were

compared.

We found very considerable damage with intensities commonly used for therapy (i.e., 1.5W/cm2 and 1W/cm2, and also for 200mW/cm2) for various periods. All the D.N.A. molecules were broken down. We are reporting this in more detail elsewhere. We thus found that at the therapeutic doses and at 10 times less, sonication is mutagenic for purified D.N.A. For doses commonly used in obstetrics (i.e. 20mW/cm2, IMHz), we found no effect. But we have to point out that a dose of 200mW/cm2, which has a drastic effect, is only 10 times the intensity used for obstetric diagnosis, which is sometimes applied for a much longer time (for several hours in fetal-heart monitoring during labour). Besides, sonication can have short peaks of higher intensities than those given by calibration of the machine. Although our results cannot lead to a conclusion as far as their in-vivo implications are concerned, we must point out that intensities which are only one order of magnitude higher than those used in obstetric diagnosis, and lower than those used in therapy, cause large-scale fragmentation of purified D.N.A. in solution. Therefore, we believe that, at present, although considerable development of the use of ultrasound in diagnosis is completely justified, exposure should be minimised, especially in obstetric use, because the growing fetus is very susceptible to mutagenic agents. As for ultrasound doses used for therapy, although they appeared to be mutagenic on purified D.N.A.., the consequences are, a priori, less important than if fetal cells are involved, although it is known that mutations can contribute to cancer aetiology. We think it is important to continue to study the

effect of ultrasound

HORMONE SENSITIVITY AND BREAST CANCER

at

medical doses

on D.N.A.

in vivo, and

we

completely agree with Fischman, who emphasises’ "that it is most important, in working with ultrasound, that all of its variables be specified and, in so far as possible, controlled". Faculté de Médicine U.L.B., 97 Rue aux Laines, 1000 Bruxelles.

Anthropagenetika, Vrije Universiteit

van

Brussel.

Clinique de Gynecologie et Obstétrique, Hôpital Universitaire Brugmann, Brussels, Belgium. 1. Fischman, H. Lancet,

1973, ii, 920.

H. GALPERIN-LEMAITRE M. KIRSCH-VOLDERS

S. LEVI.

years in the

use

of

examination of breast cancersl-3 which were modified and applied by Prof. J. R. Hobbs’s group.4-7 More recently we have extended the use of these methods to hormone bioassay,8 and our results have shown that cytochemistry can be made very precisely quantitative. However, this experience has made clear to us what meticulous attention must be paid to niceties of technique if meaningful results are to be obtained. Wilson and Carr suggest that variation in section thickness may give rise to errors in the interpretation of results. In fact, consideration of the principles underlying tissue densitometry indicates that such variation will completely invalidate the results. These principles are essentially similar to those underlying conventional spectrophotometry: in the one case, light is absorbed by a solution contained in a cuvette; in the other, by a stain precipitated in a tissue section. Thus the thickness of the section is precisely analogous to the path length of the cuvette, which, as the most junior chemical-pathology technician well knows, is directly proportional to the measured optical density. Constant section thickness is therefore absolutely essential. Unfortunately it is one thing to recognise the necessity of constant section thickness, but quite another to achieve it. Butcher9 has shown that variation in cutting speed is one cause of inconstant thickness, and we would recommend that normally a cryostat with an automatic cutting device should be used for quantitative work, as this instrument cuts with a uniform speed. Inconstant section thickness, however, cannot account for the anomaly reported by Wilson and Carr, whereby sections which appeared to the eye to differ in staining intensity showed no difference by densitometry. For, even though the concentration of stain may be equal, if one section is significantly thicker than another it will appear darker to the eye as well as to the densitometer. Similarly, if the difference is apparent optically, it must also be apparent densitometrically. If not, then something is amiss with the densitometric technique. There are two likely sources of this type of error. One is that the optical densities being measured are simply too high for the instrument to discriminate between them. Just as in spectrophotometry, in tissue densitometry there is a range of optical density above which precision declines, ultimately to be lost altogether as the percentage of light transmitted to the photomultiplier becomes negligible. Optical density must therefore be kept within a suitable range by limiting the incubation time for the enzyme reaction which leads to precipitation of the formazan.

A second source of error is the optical inhomogeneity of the formazan deposit in the tissue. Unless the diameter of the scanning beam of light which is transmitted through the tissue by the densitometer is so small that it approaches the limits of resolution of the optical microscope (i.e., about 02 µm) great inaccuracies will result.l0 11 The use of inappropriate instru-

1.

Génétique Médicale,

some

cytochemical techniques to study clinical problems, so we read with interest the letter from Miss Wilson and Dr Carr (Sept. 6, p. 465) on in-vitro testing for hormone sensitivity in breast cancer. The points they raise merit further comment. We have developed cytochemical methods for the in-vitro

Altman, F. P., Chayen, J. in The Treatment of Carcinoma of the Breast (edited by A. S. Jarrett); p. 68. Amsterdam, 1968. 2. Altman, F. P., Bitensky, L., Chayen, J., Daly, J. R. Proc. Ass. clin Biochem. 1968, 5, 115. 3. Chayen, J., Altman, F. P., Bitensky, L., Daly, J. R. Lancet, 1970, i, 868 4. Salih, H., Flax, H., Hobbs, J. R. ibid. 1972, ii, 1198. 5. Salih, H., Flax, H., Brander, W., Hobbs, J R. ibid. p. 1103 6. Flax, H., Salih, H., Newton, K. A , Hobbs, J R. ibid. 1973,i, 1204 7. DeSouza, I , Morgan, L., Lewis, V. J., Raggatt, P. R., Salih, H.. Hobbs, J R. ibid. 1974, ii, 182. 8 Symposium on Cytochemical Bioassay, Clin Endocr. 1974, 3, 303 9. Butcher, R. G. Histochemie, 1971, 28, 131. 10. Butcher, R. G ibid. 1972, 32, 171. 11Bitensky, L, Butcher, R. G., Chayen, J. in Lysosomes in Biology and Pathology (edited by J. T. Dingle); vol 3, p. 465. Amsterdam, 1973