370 patients who had carcinoma of the breast with involved axillary lymph-nodes were randomised after total mastectomy and axillary clearance to receive either no additional treatment or melphalan 6 mg/m2 daily for 5 days every 6 weeks for sixteen cycles. There was a trend towards longer relapse-free survival (RFS) in patients treated with melphalan, but this was not significant either in the whole series or in sub-groups according to menopausal status or extent of nodal involvement. In patients receiving melphalan RFS was not significantly affected by either the occurrence of amenorrhoea or the dosage of melphalan received. Overall survival did not differ significantly between the two groups. The results of this trial suggest that there is no place for the use of melphalan as adjuvant therapy in the management of early breast cancer.

Naloxone, an opioid antagonist, and meptazinol, an opioid antagonist with agonist properties, were tested in a double-blind placebo-controlled trial in 24 Ethiopian patients with louse-borne relapsing fever. The potentially fatal Jarisch-Herxheimer reaction (J-HR), which invariably follows tetracycline treatment of the disease, was unaffected by naloxone, 30-40 mg intravenously, but was diminished by meptazinol, 300-500 mg intravenously. Compared with naloxone and placebo, meptazinol reduced the clinical severity of the reaction, significantly delayed and shortened its chill phase, delayed the rise in temperature, and reduced peak temperature and changes in pulse and respiratory rates and leucocyte count. High-dose corticosteroids given before or at the time of tetracycline treatment failed to alter the reaction, which is thought to result from release of endotoxin-like substances. Meptazinol is the first effective treatment for the J-HR of louse-borne relapsing fever. This finding suggests a role for endogenous opioids in the pathogenesis of the J-HR.

The immediate haemodynamic effects of intravenous frusemide (1 mg/kg) and intravenous isosorbide dinitrate (50-200 micrograms/kg/h) were compared in a prospective, randomised, between-group study in 28 men with radiographic and haemodynamic evidence of left ventricular failure following acute myocardial infarction. The diuresis induced by frusemide reduced the left heart filling pressure and cardiac output and transiently raised systemic blood-pressure. In contrast, isosorbide dinitrate was accompanied by a reduction in systemic blood-pressure and peripheral resistance with the result that the cardiac output was not decreased despite a large fall in the pulmonary vascular and left heart filling pressures. These results indicate that reduction of excessive preload by venodilatation may be haemodynamically superior to that induced by diuresis in terms of both reducing myocardial oxygen consumption and maintaining peripheral perfusion. The influence of these contrasting treatments on the prognosis of these high-risk patients warrants further study.

There are two apparently conflicting theories on the pathogenesis of exercise-induced asthma. One view is that exercise-induced asthma is directly related to respiratory heat exchange and that mast cells are not involved. The other theory explains exercise-induced asthma on the basis of a temperature-independent release of mast-cell mediators. A theory is put forward here that airway cooling in asthmatic subjects during exercise leads directly to mast-cell degranulation and that this explains the association between exercise, respiratory heat exchange, and mediator release.

64 women with operable breast cancer were randomly allocated at the time of mastectomy to a group receiving immediate breast reconstruction or to a control group to whom breast reconstruction was offered 12 months later. The objective of the trial was to determine whether immediate breast reconstruction affected the psychosocial morbidity of mastectomy. Immediate reconstruction reduced the psychiatric morbidity assessed 3 months after operation, predominantly in women with unsatisfactory marriages. Women who underwent reconstruction had more freedom of dress and were less likely to be repulsed by their own naked appearance than women who did not undergo reconstruction. Sexual and social morbidity were not affected.

Atenolol was compared with placebo in a randomised and double-blind prospective study of 120 women with mild to moderate pregnancy-associated hypertension who were also initially managed conventionally by bed rest. Atenolol given once daily significantly reduced blood-pressure, prevented proteinuria, and reduced the number of hospital admissions. Loss of blood-pressure control leading to withdrawal from the study was commoner among the placebo group, whose babies had a high morbidity. Respiratory distress syndrome occurred only in the placebo group. Intrauterine growth retardation, neonatal hypoglycaemia, and hyperbilirubinaemia occurred with the same frequency in the two groups. Neonatal bradycardia was more common after atenolol but the systolic blood-pressure of the babies was the same in both groups. There was no difference between the groups in maternal symptoms which could have been attributed to beta-blocker therapy. Thus atenolol is more effective than conventional obstetric management in this form of hypertension and does not adversely affect mother or baby.

A 96 h infusion of epoprostenol (prostacyclin) or placebo was given as a treatment for rest pain to 28 patients with severe peripheral arterial disease. 24 h after the end of the infusion patients given epoprostenol had significantly less pain and took fewer analgesics tablets than did the controls. In some patients the epoprostenol effect lasted for over 1 month.

Two groups of outpatients (7 in each group) with Raynaud's syndrome, matched for severity of illness, were randomly allocated to receive at weekly intervals for three weeks either a 5 h intravenous infusion of buffer or epoprostenol (prostacyclin, PGI2) in buffer (7.5 ng/kg/min after the first hour). PGI2 reduced the frequency and duration of ischaemic attacks (both p less than 0.01). Hand temperature measurements with a thermocouple were significantly improved at 1 week; 6 weeks after the last infusion hand temperatures had returned to baseline. There was a corresponding loss of clinical response 8-10 weeks after the last infusion.

Tamoxifen ('Nolvadex'), an anti-oestrogen, has been evaluated as an adjunct to the local treatment of early breast cancer in a prospective randomised clinical trial. 1285 women (with pathological stage II premenopausal and pathological stage I and II postmenopausal disease) were treated by total mastectomy with either axillary node clearance or axillary node sampling and then randomised to receive either tamoxifen 10 mg twice daily for two years or no further treatment. Treatment failure (recurrent disease or death) at 21 months was reduced in patients receiving tamoxifen (14 X 2%) compared with controls (20 X 5%) (p = 0 X 01). This is equivalent to a prolongation of the disease-free interval from 21 months to 30 months at the mean follow-up time of 21 months. Subgroup analyses by menopausal, axillary lymph node, and oestrogen receptor status did not reveal a significantly different treatment effect in any of these subgroups. There has been no significant effect on mortality at this point in the study. This endocrine adjuvant therapy was well tolerated and treatment was discontinued in only 14 (2 X 2%) patients as a direct result of side-effects. Thus, tamoxifen significantly delays recurrence in early breast cancer. The magnitude of the effect is comparable with that associated with adjuvant cytotoxic chemotherapy at a similar follow-up time, but with minimal toxicity and excellent compliance.

The efficacy and safety of a single daily dose of sodium stibogluconate, 20 mg/kg body weight, given by deep intramuscular injection was compared with the conventional dose of 10 mg Sb/kg body weight in a randomised trial in Kenyan children and adults with visceral leishmaniasis. Splenic aspiration proved a safe and simple method for assessing parasitological response to treatment. In children the higher dose was associated with a faster clinical and parasitological response, and 100% were cured within 4 weeks, compared with 60% receiving the lower dosage. This difference is statistically significant by life-table analysis (x2 = 4.41, p less than 0.05). The superiority of the higher dose was not, however, seen in adults. In both children and adults the higher dose given daily for 2--4 weeks and in one patient for up to 7 weeks was found to be safe and well tolerated. It is likely, but not proven, that the use of sodium stibogluconate in a dose of 20 mg/kg bw daily for 4 weeks will reduce the relapse-rate in Kenyan children with visceral leishmaniasis.

74 insulin-dependent diabetic patients with background retinopathy were randomised to continue with usual diabetic care (group U) or to a more intensive programme (group A) using ultralente insulin as basal cover and soluble insulin at mealtimes. Group A attended the clinic more frequently, received closer dietary supervision, and were taught home blood glucose monitoring. Group A had a significantly lower mean glycosylated haemoglobin level during the study, although the mean level also fell in group U towards the end of year 2. Renal and sensory-nerve function were significantly better preserved in group A than in group U. Significant improvements were also seen in low-density-lipoprotein-cholesterol and whole-blood low-shear viscosity. The rate of progression of retinopathy was similar in both groups. It appears that a modest improvement in diabetic control, obtainable in most clinics, has been associated with a reduction in the progression of diabetic tissue damage.

30 insulin-dependent diabetic patients with background retinopathy were randomised to conventional treatment (UCT) or treatment with continuous subcutaneous insulin infusion (CSII). They were followed prospectively for 1 year with fortnightly seven-sample home blood glucose measurements and retinal examinations every 6 months. Mean blood glucose and stable haemoglobin A1c during months 3-12 were significantly lower in the CSII than the UCT group. Retinal morphology deteriorated during the year with no significant differences between UCT and CSII groups. The frequency of deterioration was highest in the CSII group, especially among the 10 patients with best glycaemic control. Proliferative retinopathy developed in 3 patients--2 of these were CSII treated. Retinal function (oscillatory potential, macular recovery time, and posterior vitreous fluorophotometry) improved significantly with CSII treatment and deteriorated significantly with UCT. Changes in retinal function were most pronounced in patients with the best and the poorest regulated glycaemic control.

Thirty-four normocalcaemic women with multiple osteolytic bone metastases from breast cancer were randomly allocated to treatment with disodium dichloromethylene diphosphonate (Cl2MDP) 1600 mg/day orally (17) or placebo (17) for 3-9 months. Fasting urinary hydroxyproline/creatinine and calcium/creatinine ratios declined in the Cl2MDP group but not in the placebo group. Four patients in the placebo group died from hypercalcaemia. New bone metastases were more common in patients on placebo and these patients also required more analgesic drugs than those on Cl2MDP. Cl2MDP seemed to reduce bone pain and bone resorption and prevent the development of hypercalcaemia caused by osteolytic metastases. The formation of new bone metastases and the growth of old ones seemed to be retarded by Cl2MDP.