Interference in thyroid-function tests in postpartum thyroiditis

Three women are described from a study of patients with postpartum thyroiditis whose sera gave spuriously increased concentrations of free thyroid hormone because of antibody binding of radiolabeled thyroxin (T4) and triiodothyronine (T3) analogs. All of the women showed increased serum concentrations of thyroid autoantibodies. The antibody binding of radiolabeled analogs and its effect on free T4 and free T3 assays disappeared by 48 weeks postpartum. Postpartum women who develop thyroid autoantibodies have approximately 2% prevalence of increased binding of radiolabeled analogs, which can result in an interference in thyroid hormone assays involving T4 and T3 analogs.

Interference in thyroid-function tests in postpartum thyroiditis

Three women are described from a study of patients with postpartum thyroiditis whose sera gave spuriously increased concentrations of free thyroid hormone because of antibody binding of radiolabeled thyroxin (T4) and triiodothyronine (T3) analogs. All of the women showed increased serum concentrations of thyroid autoantibodies. The antibody binding of radiolabeled analogs and its effect on free T4 and free T3 assays disappeared by 48 weeks postpartum. Postpartum women who develop thyroid autoantibodies have approximately 2% prevalence of increased binding of radiolabeled analogs, which can result in an interference in thyroid hormone assays involving T4 and T3 analogs.

The effect of weight correction on risk calculations for Down's syndrome screening

Recent advances in prenatal screening have led to the possibility that the risk of Down's syndrome associated pregnancy may be assessed by blood tests for maternal serum alphafetoprotein, human chorionic gonadotrophin (and possibly unconjugated oestriol) taken at 15-18 weeks of gestation. In neural tube defect screening correction of maternal serum alphafetoprotein for maternal weight has been recommended, although the precise method for weight correction is still under debate. We report an assessment of weight correction for maternal serum alphafetoprotein and human chorionic gonadotrophin based on 1408 singleton pregnancies and for unconjugated oestriol based on 197 singleton pregnancies. We demonstrate that weight correction of maternal serum alphafetoprotein and human chorionic gonadotrophin is statistically valid but that correction of unconjugated oestriol is not.

The clinical spectrum of staphylococcal bacteraemia: a review of 101 Melanesian patients from Papua New Guinea

The clinical features of 101 Melanesian patients with Staphylococcus aureus bacteraemia observed during two 2-year periods (1977-1979 and 1985-1987) in a university teaching hospital in Papua New Guinea are reviewed. The age of the patients ranged from 12 to 70 years. There were 69 males and 32 females. Diabetes mellitus, found in 15 patients, was the most common predisposing factor. Most of the patients (87%) had community-acquired infection. Soft-tissue infection, pneumonia, arthritis, osteomyelitis, intravenous-site thrombophlebitis, cerebral abscess, endocarditis and cavernous sinus thrombosis were among the clinical entities observed. Soft tissues and lungs were the most common sites of primary and secondary foci of infection, respectively. All but 1 of the 101 blood isolates were resistant to penicillin G and none was resistant to methicillin. The overall case fatality rate was 24%. These data demonstrate that staphylococcal bacteraemia in adult Papua New Guineans is mostly community acquired and has a high mortality. Skin and soft tissues are the major primary foci of infection leading to staphylococcal bacteraemia.

A long-term follow-up of postpartum thyroiditis

To investigate the long-term outcome of postpartum thyroiditis (PPT), 43 patients with PPT and 171 control women were evaluated 3.5 (range 2-4) years postpartum. Ten (23%) PPT patients were hypothyroid compared to none of the controls (P less than 0.001). Factors associated with the development of hypothyroidism were high antimicrosomal antibody titre measured at 16 weeks gestation (P less than 0.01), severity of hypothyroid phase of PPT, multiparity, and a previous history of spontaneous abortion. The presence of microsomal antibody but no PPT in one pregnancy did not prevent the occurrence of PPT in the next pregnancy in two patients and a further five patients had PPT in two successive pregnancies. There was no association between HLA haplotype, family history of thyroid disease, smoking or frequency of oral contraception, and the development of long-term hypothyroidism after PPT. It is concluded that permanent hypothyroidism is an important sequel to PPT and patients with PPT should be followed up appropriately.

Concentrations of free thyroxin and free triiodothyronine in serum of patients with thyroxin- and triiodothyronine-binding autoantibodies

Between 1982 and 1989 we identified 47 subjects with spuriously increased concentrations of free thyroxin (FT4) or free triiodothyronine (FT3) related to autoantibody interference in analog FT4 and (or) FT3 methods. The incidence of autoantibody interference observed during one year (1988) was 1 in 2460. In the subjects identified, 51% and 11%, respectively, showed an increased binding of radiolabeled T4 or T3 analog alone; 38% had an increased binding of both. Of 36 patients tested, 71% had autoantibodies to thyroglobulin and microsomal fraction of the thyroid, 19% to microsomal fraction alone, and 9.5% to thyroglobulin alone. In eight subjects, spuriously increased FT4 concentrations were reported with the following FT4 methods (in decreasing order of interference): Coat-A-Count, Amerlex-M, Amerlite, Seria, Magic Lite, Amerlex-MAB. In the same eight subjects, Amerlex-M and Seria reported spuriously increased concentrations of FT3.

Concentrations of free thyroxin and free triiodothyronine in serum of patients with thyroxin- and triiodothyronine-binding autoantibodies

Between 1982 and 1989 we identified 47 subjects with spuriously increased concentrations of free thyroxin (FT4) or free triiodothyronine (FT3) related to autoantibody interference in analog FT4 and (or) FT3 methods. The incidence of autoantibody interference observed during one year (1988) was 1 in 2460. In the subjects identified, 51% and 11%, respectively, showed an increased binding of radiolabeled T4 or T3 analog alone; 38% had an increased binding of both. Of 36 patients tested, 71% had autoantibodies to thyroglobulin and microsomal fraction of the thyroid, 19% to microsomal fraction alone, and 9.5% to thyroglobulin alone. In eight subjects, spuriously increased FT4 concentrations were reported with the following FT4 methods (in decreasing order of interference): Coat-A-Count, Amerlex-M, Amerlite, Seria, Magic Lite, Amerlex-MAB. In the same eight subjects, Amerlex-M and Seria reported spuriously increased concentrations of FT3.

Rate of cerebral embolic events in relation to antibiotic and anticoagulant therapy in patients with bacterial endocarditis

To quantitate embolic risk we studied a retrospective series of 61 patients with strictly defined bacterial endocarditis. Eighteen patients had neurological complications attributable to cerebral embolism. Seventeen embolic episodes occurred prior to antibiotic treatment and 8 episodes after its commencement. The rate of embolism per patient-week during a 20-week observation period showed a highly significant almost 4-fold reduction to a low level soon after antibiotics were started. Patients already on anticoagulation for prosthetic valves had the same embolic risk as those not so treated. The evidence suggests that anticoagulation at the time of diagnosis of bacterial endocarditis is not warranted.

The epicardial electrogram: a quantitative assessment during balloon angioplasty incorporating monophasic action potential recordings

An electrogram was recorded from the angioplasty catheter guide wire when coronary blood flow was interrupted in 20 patients undergoing percutaneous transluminal coronary angioplasty. Monophasic action potentials were recorded from the right ventricular septum together with the routine electrocardiogram. The patients were studied during angioplasty for lesions in the left anterior descending (12), circumflex (3), and right coronary arteries (6). ST elevation in the electrogram recorded in the left anterior descending and circumflex systems was usually more obvious than that in the electrocardiogram. Signals obtained from the right coronary artery were of very low amplitude and registered only minimal ST changes. The ST elevation developed in the electrogram during insertion of the catheter before inflation of the balloon in 11 of the 15 patients undergoing angioplasty of the left system. In eight of the patients showing pre-inflation ST elevation the ST shift lessened after successive inflations. Monophasic action potential recordings were obtained during 45 balloon inflations in 19 patients. In those patients undergoing angioplasty for lesions of the circumflex coronary artery the monophasic action potential showed no change during balloon inflation. In patients undergoing angioplasty for the right coronary artery the mean normalised duration at 60 seconds' occlusion was 99.6 (1.5)% of control. Of a total of 25 occlusions in the patients undergoing angioplasty for the left anterior descending coronary artery 19 showed shortening of less than 5%, five showed shortening between 5 and 10%, and one showed a shortening of 16.4% in the monophasic action potential. The QT interval was satisfactorily measured in the electrogram during 36 balloon inflations, and in 24 of these it was also measured in the electrocardiogram. QT changes in the electrogram tended to be the opposite of those in the electrocardiogram. When changes in RR interval were minimal (less than 20 ms) during the balloon inflation 14 of 17 electrograms showed QT prolongation but only one of 12 electrocardiograms showed prolongation. Conversely one of 17 electrograms showed shortening compared with eight of 12 electrocardiograms. There was angiographic evidence of the development of collaterals in six of 15 patients undergoing angioplasty of the left system. ST segment elevation in both the electrogram and electrocardiogram was less pronounced in these patients than in those without evidence of the development of collaterals. ST segment changes recorded from the angioplasty guide wire provide a more sensitive index of ischaemia than the surface 12 lead electrocardiogram, and fall in ST segments on balloon deflation is a prognostic index of a good angiographic result in the left anterior descending and circumflex arteries, but not in the right coronary artery.

Growth hormone responses to growth hormone releasing factor in primary degenerative dementia

The growth hormone (GH), thyroid-stimulating hormone (TSH), and prolactin (PRL) responses to growth hormone releasing factor (GRF) were investigated in 18 patients suffering from primary degenerative dementia (PDD) and in 20 age- and sex-matched normal elderly controls. There was no significant difference in the growth hormone response to GRF stimulation between patients and controls, and in neither subject group was there a demonstrable TSH or prolactin response to GRF. These findings indicate that the pathophysiology underlying the blunted growth hormone response to pharmacological challenge in PDD must lie at a suprapituitary level.

Antibody interference in a two-site immunometric assay for thyrotrophin

Of 1585 consecutive serum samples referred for thyroid function testing, 14 gave erroneously high values from a two-site immunoenzymometric assay for thyrotrophin. The addition of mouse or newborn calf serum to the assay failed to correct the interference. In serum samples from 11 patients who were available for follow up a higher concentration of mouse, but not of horse, sheep or rabbit serum reduced the interference. The interference was associated only with assay systems which employed horseradish peroxidase but not 125I as a label. Addition of mouse serum and anti-IgM to the assay reagents successfully removed the interference.

Effect of nadolol on plasma lipids in hyperthyroidism

The effect of Nadolol treatment on lipid subfractions in a group of 23 hyperthyroid patients was assessed in a randomised double-blind placebo controlled trial lasting six weeks, carbimazole being given to both groups from weeks 2 to 6. Clinical and biochemical euthyroidism was seen in both groups at 6 weeks; no effect of nadolol on peripheral monodeiodination of T4 to T3 was observed. At time 0 there were significant negative correlations between total cholesterol and free T3 (r = 0.68), and free T4 (r = 0.54). In the Nadolol group there were significant rises between 0 and 6 weeks in total cholesterol (52.6%, P less than 0.01), LDL cholesterol (30.3%, P less than 0.01) and HDL cholesterol (18.2%, P less than 0.05). HDL cholesterol rose significantly in the placebo group (12.4%, P less than 0.05) but there were no significant increases in LDL cholesterol or total cholesterol. The rise in triglyceride during this period in the Nadolol group (64.7%, P less than 0.05) was significantly greater (P less than 0.05) than the rise in the placebo group (8.8%). Nadolol increases triglyceride more than placebo during the early management of hyperthyroidism.

The clinical value of serum prolactin measurement in the differential diagnosis of complex partial seizures

The time course of changes in serum prolactin after complex partial seizures has been determined and compared to similar changes after other types of seizure and non-epileptic attacks. Seizures in 33 subjects were recorded on video EEG telemetry. Peak serum prolactin concentrations occurred 15-20 min after tonic-clonic seizures, 10 min after complex partial seizures, and were highest after generalised tonic-clonic seizures. Serum prolactin concentrations remained less than 1000 mU/l after absences and non-epileptic attacks. Application of Bayes' theorem showed that where serum prolactin was greater than 1000 mU/l 5-10 min post event this would identify genuine tonic-clonic or complex partial seizures. The false negative rate of this test was 9% for tonic-clonic seizures and 38% for complex partial seizures. Failure of serum prolactin to rise after an attack is of little value in distinguishing complex partial seizures from non-epileptic attacks.

GH responses to growth hormone releasing factor in depression

The growth hormone (GH), thyrotrophin (TSH) and prolactin response to growth hormone releasing factor (GRF) was investigated in 18 patients suffering from major depression with melancholia and in 18 age- and sex-matched normal controls. There was no significant difference in the GH response to GRF stimulation between the patients and controls and in neither subject group was there a demonstrable TSH or prolactin response to GRF. These findings indicate that the pathophysiology underlying the blunted GH response to pharmacological challenge, demonstrated in other studies, must lie at a suprapituitary level.

Evaluation of a new strategy for detection of thyroid dysfunction in the routine laboratory

We assessed the use of a new strategy for detecting thyroid disorders, utilizing a sensitive assay for concentrations of thyrotropin (TSH) and free thyroid hormone in serum as follow-up tests. Of 1279 patients who were not on thyroxin (T4) replacement treatment, 82% could be classified as euthyroid and would require no further tests. In patients who were on T4 replacement, 41% fell into the euthyroid category and would require no further tests. Using this strategy to replace our existing strategy of free thyroxin as a "first-line" test would reduce the proportion of patients who would require one or more follow-up tests from 49% to 24%.

Evaluation of a new strategy for detection of thyroid dysfunction in the routine laboratory

We assessed the use of a new strategy for detecting thyroid disorders, utilizing a sensitive assay for concentrations of thyrotropin (TSH) and free thyroid hormone in serum as follow-up tests. Of 1279 patients who were not on thyroxin (T4) replacement treatment, 82% could be classified as euthyroid and would require no further tests. In patients who were on T4 replacement, 41% fell into the euthyroid category and would require no further tests. Using this strategy to replace our existing strategy of free thyroxin as a "first-line" test would reduce the proportion of patients who would require one or more follow-up tests from 49% to 24%.

Postpartum thyroid dysfunction in Mid Glamorgan

A high prevalence of postpartum thyroid dysfunction has been reported in several countries, but there have been no systematic studies of its prevalence in Britain. Among a group of 901 consecutive, unselected pregnant women thyroid autoantibodies were detected in 117 (13%) at booking. The clinical course of postpartum thyroid dysfunction, factors associated with its development, and its likely prevalence were defined in 100 of these women with thyroid antibodies and 120 women with no such antibodies who were matched for age. None of the women had a history of autoimmune thyroid disease. Normal reference ranges for thyroid function during pregnancy and post partum were established in the 120 women negative for thyroid antibodies. On the basis of these observations postpartum thyroid dysfunction was observed in 49 (22%) of the 220 women studied, and the prevalence in the total group of 901 women was estimated to be 16.7%. Thyroid dysfunction, mainly occurring in the first six months post partum, was usually transient and included both destruction induced hyperthyroidism and hypothyroidism. The development of the syndrome was significantly related to smoking more than 20 cigarettes a day and the presence of thyroid microsomal autoantibodies at booking. Of the 16 women with a family history of thyroid disease in whom thyroid microsomal autoantibody activity was detectable at booking, 11 developed thyroid dysfunction. Age, parity, presence of goitre at presentation, duration of breast feeding, and the sex and birth weight of the infant were not associated with the development of postpartum thyroid dysfunction. The mood changes experienced by women post partum may in part be associated with altered thyroid function during this time.

The effect of preoperative Lugol's iodine on thyroid blood flow in patients with Graves' hyperthyroidism

A study was conducted to investigate the effect of Lugol's iodine on the superior thyroid artery (STA) blood flow with use of a Duplex ultrasound scanner for 12 patients with Graves' disease. All patients were treated with antithyroid drugs until they were euthyroid and then, with randomization, the patients received either Lugol's iodine, 0.3 ml thrice daily, or placebo for 9 days in a double-blind fashion. Antithyroid drugs were continued throughout the study. Reduction in the diameter, time-averaged velocity (TAV), and volume flow (VF) of the STAs was demonstrated in all patients in the treatment group, whereas there were no consistent trends in the placebo group. The changes in TAV and VF were significantly different between the placebo and treatment groups (p less than 0.01 for TAV and p less than 0.005 for VF). These changes were more marked in patients with high initial VF and minimal in patients with low initial VF. On the basis of these results, we recommend that patients with high thyroid blood flow before thyroidectomy should receive Lugol's iodine preoperatively.

Determinants of free theophylline clearance in asthma

1. The free and total plasma and saliva concentrations of theophylline were measured during a dosing interval at steady state in nineteen asthmatic subjects receiving a once-daily theophylline preparation (Riker TCR-1). 2. Saliva theophylline clearance (mean +/- s.d. 5.8 +/- 2.1l h-1) was closely related to total plasma theophylline clearance (mean 3.6 +/- 1.2l h-1) (r = 0.958, n = 19, P less than 0.001). 3. Saliva theophylline clearance was closely related (r = 0.967, n = 19, P less than 0.001) and numerically very similar to the free plasma theophylline clearance (mean 5.8 +/- 1.9l h-1) (mean difference = 0.06 +/- 0.12 s.e. mean). 4. Free plasma theophylline clearance was significantly, although weakly, related to body weight and to the plasma free thyroxine concentration which together accounted for over 40 per cent of the variability in free clearance. 5. Theophylline administered for 2 weeks did not affect plasma free thyroxine (FT4) free tri-iodothyronine (FT3) or reverse tri-iodothyronine (rT3) concentrations compared with placebo.

Non-adenomatous inappropriate TSH hypersecretion and euthyroidism requires no treatment

The syndrome of inappropriate TSH secretion is described in a euthyroid girl and her father. Based on nuclear T3 binding studies in fibroblasts, generalized tissue resistance was associated with a lower binding affinity for T3 in nuclear extracts suggestive of a structurally abnormal receptor for T3. Early recognition of the syndrome and observation of the short-term response to thyroid medication prevented unnecessary trials of antithyroid medication and later radical ablative thyroid treatment.

Serum free thyroxine and free triiodothyronine concentrations in healthy fullterm, preterm and sick preterm neonates

There are few data available on free thyroid hormone concentrations in the early neonatal period. With the widespread application of screening procedures for detecting congenital hypothyroidism there is a need for reference ranges in neonates. In this study we have evaluated thyroid function in healthy fullterm and preterm neonates, and sick neonates all within one to 10 days postnatal age. Our data indicates that free thyroxine but not free triiodothyronine is higher in fullterm neonates than the adult reference range and that both free thyroid hormone concentrations are reduced in healthy and sick preterm neonates as compared to fullterm neonates. Assessment of thyroid function in the early neonatal period needs to take into account these changes particularly in preterm and sick preterm neonates.

Thyroid status in senile dementia of the Alzheimer type (SDAT)

Thyroid function was investigated in a group of 21 patients with severe senile dementia of the Alzheimer type (SDAT) and in a group of 17 age and sex matched normal controls. Free thyroid hormone levels (triiodothyronine (T3) and thyroxine (T4) were measured, as were also the thyrotrophin (TSH), prolactin (PRL) and growth hormone (GH) responses to thyrotrophin releasing hormone (TRH)). When compared to controls, patients demonstrated a significantly lower free T3 value (but not free T4), a blunted TSH response to TRH, slightly elevated basal PRL and GH values and a small GH response to TRH. However, all differences were small in biological terms and were within the laboratory's normal range. This emphasizes the relative normality of neuroendocrine function, particularly thyroid status, in SDAT.

An enhanced serum prolactin response to TRH in the presence of GABA transaminase inhibition

Patients with epilepsy were found to have an increased 20 minute prolactin response to intravenous TRH stimulation when receiving the GABA-T inhibiting drug vigabatrin. Enhanced GABA activity may either reduce basal prolactin levels whilst allowing a normal pituitary response to TRH stimulation, or may overcome the inhibitory effects of dopamine on pituitary prolactin release.

Clinical value of immunoradiometric assay of thyrotropin for patients with nonthyroidal illness and taking various drugs

Using a two-site immunoradiometric assay, we measured concentrations of thyrotropin (TSH) in serum of 134 clinically euthyroid subjects, 93 patients with nonthyroidal illness, and 80 patients who were being treated with various drugs. Abnormal concentrations of TSH, free thyroxin, and free triiodothyronine, respectively, were recorded in serum of three (3.2%), 19 (20.4%), and 37 (39.8%) of the patients with nonthyroidal illness and in three (3.8%), five (6.3%), and 10 (12.5%) of the patients taking drugs. TSH could be detected in all patients' serum samples. We conclude that, for most patients without thyroid disease, a basal (i.e., unstimulated) measurement of their TSH concentration in serum will indicate their thyroid status more reliably than will assay of free thyroxin or free triiodothyronine.

Clinical value of immunoradiometric assay of thyrotropin for patients with nonthyroidal illness and taking various drugs

Using a two-site immunoradiometric assay, we measured concentrations of thyrotropin (TSH) in serum of 134 clinically euthyroid subjects, 93 patients with nonthyroidal illness, and 80 patients who were being treated with various drugs. Abnormal concentrations of TSH, free thyroxin, and free triiodothyronine, respectively, were recorded in serum of three (3.2%), 19 (20.4%), and 37 (39.8%) of the patients with nonthyroidal illness and in three (3.8%), five (6.3%), and 10 (12.5%) of the patients taking drugs. TSH could be detected in all patients' serum samples. We conclude that, for most patients without thyroid disease, a basal (i.e., unstimulated) measurement of their TSH concentration in serum will indicate their thyroid status more reliably than will assay of free thyroxin or free triiodothyronine.

Multivariate analyses of the EEG in normal adolescents

We investigated the computerized electroencephalography (cEEG) in a group of 85 normal adolescents. The spatial frequency of the cEEG was analyzed using various multivariate statistical techniques. The factor analyses revealed three factors that accounted for most of the variability in the data and that could be interpreted as a low versus high frequency, a beta frequency, and a frontal versus posterior factor. The cluster analysis revealed various EEG patterns that distinguished among subgroups of normal individuals.

The effect of nadolol on heart rate in hyperthyroidism. A controlled trial

Twenty hyperthyroid patients were randomly assigned in a double-blind fashion to receive either nadolol 80 mg/day or placebo for 2 weeks; all patients then took carbimazole as well from 2-6 weeks. Twenty-four hour Holter ECG recordings at 0, 2 and 6 weeks showed that nadolol reduced the mean maximum heart rate by 19.9% (P less than 0.0005) at 2 weeks and by 30.3% (P less than 0.0005) at 6 weeks compared to 5.2% (ns) and 18.3% (P less than 0.0005) in patients taking placebo. There was no alteration of the normal circadian variation of heart rate by nadolol. The minimum heart rate before therapy was significantly correlated with FT4 (r = 0.52) and with FT3 (r = 0.44). The percentage of time per hour during which the heart rate was greater than 100 was reduced by 79% at week 2 by nadolol compared to 22% in the placebo group. At the 6 week point the placebo group still had a tachycardia (mean maximum heart rate 101.6 beats/min +/- 15.2 SD) compared to the nadolol group (80.4 +/- 7.7). Nadolol did not cause excessive bradycardia. It is effective in the early management of hyperthyroidism and should be given for at least the first 4-6 weeks.

Enhanced luminescence immunoassay: evaluation of a new, more sensitive thyrotropin assay

We measured thyrotropin (TSH) with an enhanced luminometric assay ("Amerlite"; Amersham International). The detection limit of the assay is 0.02 milli-int. unit/L. Within-assay precision was 6.7 and 7.8% at 3.77 and 12.1 milli-int units/L, respectively, and between-assay precision was almost identical, whether singleton or duplicate samples were assayed. TSH measured in 132 euthyroid subjects ranged from 0.06 to 4.13 milli-int. units/L (mean 1.52, SD 0.86). Similar concentrations were found in 20 healthy pregnant women and 19 of 20 healthy post-menopausal women (one of whom had undetectable TSH). In 17 patients with primary hypothyroidism, TSH concentrations ranged from 9.34 to greater than 200 milli-int. units/L; and in 53 of 59 patients with hyperthyroidism, TSH concentrations were undetectable, ranging in the remaining six from 0.03 to 0.06 milli-int. unit/L. Results for TSH in 28 patients stimulated with thyroliberin were consonant with the results of the thyroliberin test in 25 cases. Thus, for most patients, measurement of a basal TSH concentration evidently will predict their thyroidal status and also the response to thyroliberin, but a few will require additional tests of thyroid function.

Enhanced luminescence immunoassay: evaluation of a new, more sensitive thyrotropin assay

We measured thyrotropin (TSH) with an enhanced luminometric assay ("Amerlite"; Amersham International). The detection limit of the assay is 0.02 milli-int. unit/L. Within-assay precision was 6.7 and 7.8% at 3.77 and 12.1 milli-int units/L, respectively, and between-assay precision was almost identical, whether singleton or duplicate samples were assayed. TSH measured in 132 euthyroid subjects ranged from 0.06 to 4.13 milli-int. units/L (mean 1.52, SD 0.86). Similar concentrations were found in 20 healthy pregnant women and 19 of 20 healthy post-menopausal women (one of whom had undetectable TSH). In 17 patients with primary hypothyroidism, TSH concentrations ranged from 9.34 to greater than 200 milli-int. units/L; and in 53 of 59 patients with hyperthyroidism, TSH concentrations were undetectable, ranging in the remaining six from 0.03 to 0.06 milli-int. unit/L. Results for TSH in 28 patients stimulated with thyroliberin were consonant with the results of the thyroliberin test in 25 cases. Thus, for most patients, measurement of a basal TSH concentration evidently will predict their thyroidal status and also the response to thyroliberin, but a few will require additional tests of thyroid function.

Evaluation of an enhanced luminescence assay for alpha-fetoprotein

We evaluated a two-site enhanced luminescence immunoenzymometric assay (Amerlite; Amersham International) for alpha-fetoprotein (AFP) in maternal serum and amniotic fluid. The assay is rapid, involving two incubations totalling 4 h. The working range of the assay for serum AFP is 5.5 to 750 kilo-int. units/L (CV less than 10%), with a sensitivity of detection of 0.2 kilo-int. unit/L. The regression equation for the Amerlite assay (y) and our in-house RIA procedure (x) was y = 0.816x + 2.9 (n = 142, r = 0.96). Analytical recovery of added AFP (code 72/227) at three concentrations was 86.7%. Serum AFP concentrations were measured at 15 to 18 weeks of gestation in subjects with normal pregnancies and in subjects whose pregnancies resulted in open neural tube defects; all of the latter had serum AFP concentrations greater than 2.5 multiples of the median. We find the Amerlite system to be an efficient, reliable system for screening for open neural tube defects without use of hazardous radioactive labels.

Thyroid function in patients with hyperprolactinaemia: relationship to dopaminergic inhibition of TSH release

It has been reported recently that patients with hyperprolactinaemia may develop hypothyroidism as a consequence of the increased inhibition of TSH release by dopamine which occurs in the majority of such patients. In this study we have evaluated thyroid function in a large number of hyperprolactinaemic patients in order to delineate more precisely the relationship between thyroid status, free thyroid hormone levels and the control of TSH release by dopamine. Biochemical euthyroidism was present in the majority of the hyperprolactinaemic patients. Our data indicate that the increased dopaminergic inhibition of TSH release does not lead to hypothyroidism. Instead, the slightly elevated basal TSH levels and TSH responses to TRH (within the normal range) may reflect the operation of a compensatory mechanism to maintain euthyroidism in the face of te increased inhibition of TSH release by hypothalamic dopamine.

Growth hormone responses to GRF 1-29 in patients with primary hypothyroidism before and during replacement therapy with thyroxine

It is well known that hypothyroidism is frequently associated with impaired GH responses to different stimuli. In the present study we have evaluated GH responses to GH-releasing factor (GRF) in patients with primary hypothyroidism before and during T4 replacement therapy. Fourteen patients (age range 26-60 years) underwent two GRF tests (1 microgram/kg) before and during replacement therapy (150 micrograms/d). Administration of T4 increased peak GH responses to GRF in 9 patients and in the group as a whole (mean +/- SEM, 17.0 +/- 2.8 vs 32.6 +/- 5.7 mU/l, P less than 0.02). When the data are analysed by means of area under the curve (AUC), the GH response to GRF was increased by T4 in 10 patients and in the group as a whole (mean +/- SEM, 51.7 +/- 14.3 vs 101.5 +/- 28.1, P less than 0.02). These data indicate that thyroid hormone replacement therapy enhances the responsiveness of the somatotroph to GRF 1-29 in patients with primary hypothyroidism.

Congenital anomalies associated with hypothyroidism

Seven of the 34 infants identified through the Welsh Hypothyroid Screening Programme have additional congenital abnormalities. Two infants have a previously undescribed syndrome, two have chromosomal abnormalities, two have congenital heart disease, and one has a myelomeningocoele. Congenital hypothyroidism often seems to be associated with other congenital abnormalities.

The preoperative and postoperative investigation of TSH and prolactin release in the management of patients with hyperprolactinaemia due to prolactinomas and nonfunctional pituitary tumours: relationship to adenoma size at surgery

We report here our results of the pre- and post-operative assessment of prolactin and TSH status in 41 hyperprolactinaemic patients who underwent pituitary surgery over a 5 year period. Preoperatively in patients with prolactinomas (n = 33) the TSH response to domperidone decreased with increasing adenoma size. When the data are expressed on a group mean basis the exaggerated TSH response to domperidone in preoperative prolactinoma patients was reduced significantly in patients rendered normoprolactinaemic by surgery but persisted in those who remained hyperprolactinaemic. Similarly the reduced preoperative PRL responses to domperidone and TRH were significantly increased by successful surgery. In contrast patients with stalk-compression hyperprolactinaemia (n = 6) due to larger lesions which were not prolactinomas all showed reduced or absent TSH responses to domperidone. The PRL responses to domperidone and TRH were reduced or absent both in patients with prolactinomas and in those with stalk-compression hyperprolactinaemia. All patients with stalk-compression hyperprolactinaemia showed a delayed pattern of TSH response to TRH with 60 min values being greater than 20 min ones. In contrast a normal pattern of TSH response to TRH was observed in all patients with hyperprolactinaemia due to prolactinomas. Postoperatively TSH and PRL responses were largely unchanged in patients with stalk-compression hyperprolactinaemia regardless of whether normoprolactinaemia was restored by surgery. In conclusion a reduced or absent PRL response to TRH or domperidone is not diagnostic of the presence of a prolactinoma since it occurs in hyperprolactinaemic patients with prolactinomas or stalk-compression. In contrast, the TSH response to acute dopamine antagonism is exaggerated in most patients with small prolactinomas but not in those with stalk-compression hyperprolactinaemia and we have found this to be helpful diagnostically since the presence of an exaggerated TSH response to dopamine antagonism is evidence against the presence of stalk-compression hyperprolactinaemia. The observation of a delayed TSH response to TRH in a hyperprolactinaemic patient should alert the clinician to the possibility of stalk-compression hyperprolactinaemia due to a large lesion which may not be a prolactinoma.

Prediction of outcome in Graves' disease after carbimazole treatment

In a prospective study to determine which factors would predict remission or relapse, 65 patients with hyperthyroid Graves' disease were treated for six months with a blocking replacement regimen of carbimazole, 40 mg daily, and triiodothyronine (T3). They were followed for one year after stopping treatment, by which time 32 (49 per cent) had relapsed. Although the treatment protocol, relapse rate and frequency of the HLA-DR3 antigen in this population were similar to those of a regionally separate Graves' population investigated previously, the predictive value of HLA-DR3 status together with thyroid stimulating antibody (TSAB) levels was strikingly different. In the present study there was no significantly abnormal distribution of any HLA antigen in the relapse group compared with those patients who achieved remission. Thyroid stimulating antibodies were detected in 62 patients (95 per cent) and fell significantly (p less than 0.05) after carbimazole treatment, irrespective of DR3 status or outcome; TSAB levels only became undetectable in nine patients (28 per cent) who subsequently relapsed and in nine patients (30 per cent) who maintained remission. T3-suppressed 20 min 123I uptake fell equally after treatment in the relapse and remission groups but continued to fall thereafter in the group which maintained remission. In these patients, 123I uptake was significantly lower at the end of the study period than at the end of treatment (p less than 0.05). Serum free T4 levels were higher before treatment in the patients who later relapsed than in those whose disease remitted (p less than 0.02). This proved the only significant marker associated with outcome but was of little predictive value in any patient. This study highlights the problem in predicting the outcome of antithyroid drug treatment, since even within the same country under similar conditions, divergent results have been obtained. It appears that the loci controlling the immune response in Graves' disease are likely to include genes lying outside the HLA-DR region. The results also suggest that the immunological effects of antithyroid drugs are maintained after stopping treatment in those patients whose disease remits.

Diagnostic value of thyrotrophin releasing hormone tests in elderly patients with atrial fibrillation

A prospective study was carried out to compare clinical and biochemical thyroid states with responses of thyroid stimulating hormone (TSH) to thyrotrophin releasing hormone (TRH) in elderly patients with either atrial fibrillation (n = 75; mean age (SD) 79.3 (6.0) years) or sinus rhythm (n = 73; mean age 78.4 (5.6) years) admitted consecutively to the department of geriatric medicine. No patient in either group had symptoms or signs of hyperthyroidism. Overall, the TSH responses to TRH did not differ significantly between the two groups. Ten (13%) of the patients with atrial fibrillation (of whom four had raised thyroid hormone concentrations) and five (7%) of the patients with sinus rhythm showed no TSH response to TRH while 26% of each group (20 and 19 patients, respectively) showed a much reduced response. Only one of 13 patients with apparently isolated atrial fibrillation showed no TSH response to TRH, and none of these 13 patients was hyperthyroid. In particular, three patients (two with atrial fibrillation and one with sinus rhythm) who showed no TSH response to TRH at presentation exhibited a return of TSH response to TRH at follow up six weeks later. In conclusion, reduced or absent TSH responses to TRH are common in sick elderly patients whether they have atrial fibrillation or sinus rhythm and whether they are euthyroid or hyperthyroid biochemically. An absence of response is therefore an uncertain marker of hyperthyroidism in these groups of patients, and diagnosis and ablative treatment should be based at least on the presence of raised circulating free triiodothyronine or free thyroxine concentrations, or both.

An automated immunoradiometric assay for human thyrotropin

In this two-site immunoradiometric assay for thyrotropin, developed for use in the "Kemtek 3000" automated radioimmunoassay system, commercially available monoclonal antibody to thyrotropin is labeled with 125I, and the solid-phase antibody is an IgG fraction of sheep antiserum to thyrotropin, covalently coupled to reprecipitated aminocellulose. There are two incubations, totalling 3 h, the sensitivity is 0.03 milli-int. unit/L. The mean thyrotropin value for 82 healthy euthyroid subjects was 1.7 milli-int. units/L (range 0.4-3.6). For 19 overtly clinically and biochemically hyperthyroid subjects the values ranged from undetectable to 0.2 milli-int. unit/L. In this assay, euthyroid and hyperthyroid subjects can be distinguished with assay of a single basal sample. The assay appears suitable for routine use as a first-line test of thyroid function.

An automated immunoradiometric assay for human thyrotropin

In this two-site immunoradiometric assay for thyrotropin, developed for use in the "Kemtek 3000" automated radioimmunoassay system, commercially available monoclonal antibody to thyrotropin is labeled with 125I, and the solid-phase antibody is an IgG fraction of sheep antiserum to thyrotropin, covalently coupled to reprecipitated aminocellulose. There are two incubations, totalling 3 h, the sensitivity is 0.03 milli-int. unit/L. The mean thyrotropin value for 82 healthy euthyroid subjects was 1.7 milli-int. units/L (range 0.4-3.6). For 19 overtly clinically and biochemically hyperthyroid subjects the values ranged from undetectable to 0.2 milli-int. unit/L. In this assay, euthyroid and hyperthyroid subjects can be distinguished with assay of a single basal sample. The assay appears suitable for routine use as a first-line test of thyroid function.

Effect of timolol on clinical features and echocardiographic assessment of left ventricular function in hyperthyroidism

The effect of timolol, a beta-adrenoceptor blocking drug on the clinical status, thyroid status and left ventricular function as measured by serial M-mode echocardiographic recordings was assessed in a double-blind randomised study in 18 hyperthyroid patients. A significant clinical improvement was documented after 2 weeks of timolol treatment compared with placebo. There was no evidence that timolol impaired peripheral monodeiodination of thyroxine (T4). There were significant increases in left ventricular fractional shortening (Fr. Sh.) and velocity of circumferential shortening (Vcf) as well as a significant decrease in the left ventricular systolic internal dimension (LVIDs) (all P less than 0.01) in the untreated thyrotoxic patients compared with a normal euthyroid control group. After timolol treatment (2/52) there were significant increases in LVIDs and LVIDd and a significant decrease in Vcf (all P less than 0.05). No further changes occurred after a further 2/52 treatment with carbimazole. The cardiac data suggest that both an augmented sympathetic drive and a direct effect of thyroid hormone on myocardial contractility are mediators of the haemodynamic changes in hyperthyroidism.