In women drawn from a randomised, population-based trial of mammography screening for breast cancer the overall actuarial survival and disease-free survival for up to 7 years of observation were similar in 94 patients with interval cancers and 178 control patients with cancer diagnosed independent of screening. The similarity was confirmed by multiple regression analysis. These results contradict the hypothesis that a high growth rate of the primary tumour is associated with a greater metastatic potential and do not support the concept that patients with interval cancers should receive more aggressive local or systemic treatment.

In a retrospective, double-blind study, paraffin-embedded renal biopsy sections from 44 children with haematuria were examined to see whether a mouse monoclonal antibody (MCA-P1) against glomerular basement membrane (GBM) could identify a subgroup of patients with hereditary glomerulonephritis (Alport's syndrome) in whom the Goodpasture antigen was abnormal. There was strong linear binding of MCA-P1 to GBM in all of 29 patients with no evidence of hereditary nephritis and in 2 patients in whom the diagnosis of hereditary nephritis was thought possible but not definite on available clinicopathological evidence. In contrast, 12 of 13 patients with strong evidence of hereditary nephritis showed no binding (9) or greatly reduced binding (3). These findings suggest that a major subgroup of patients with hereditary glomerulonephritis have biochemical abnormalities of the GBM involving Goodpasture antigen and that MCA-P1 is useful in the differential diagnosis of children with haematuria.

Total lymphoid irradiation (TLI; 1980 cGy) or sham irradiation was given to 40 patients with chronic progressive multiple sclerosis (MS) in a prospective, randomised, double-blind study. During mean follow-up of 21 months, MS patients treated with TLI had less functional decline than sham-irradiated MS patients (p less than 0.01). A significant relation was noted between absolute blood lymphocyte counts in the first year after TLI and subsequent course, patients with higher lymphocyte counts generally having a worse prognosis (p less than 0.01). TLI was well tolerated and associated with only mild short-term, and to date, long-term side-effects.

150 children aged between 2 and 9 years (mean 5.25 years) with chronic bilateral otitis media with effusion were randomly allocated to three groups who, in addition to unilateral myringotomy and grommet insertion, would have an adenotonsillectomy, an adenoidectomy, or neither. The contralateral ear was assessed. Otoscopic, impedance, and audiometric findings were recorded preoperatively and at six and twelve months postoperatively. Subtraction of the no-surgery results indicates that at twelve months adenoidectomy produces otoscopic resolution of OME in 41.7% and no-peak/peak conversion of impedance in 29.8%. The hearing gain from adenoidectomy alone was similar to that resulting from insertion of a grommet alone, but the children having adenoidectomy required fewer reinsertions to maintain adequate subjective hearing thresholds in the treated ear (26% versus 54%). Tonsillectomy conferred no additional benefit. Without treatment there was a small but significant improvement in all indices at twelve months but not at six months postoperatively.

This paper gives the first interim results of a randomised clinical trial of several drug regimens for the treatment of chronic lymphocytic leukaemia. Patients were classified as being in stage A, B, or C depending on clinical findings and peripheral blood picture. Good prognosis patients--ie, stage A (n = 455)--were randomised to receive either no treatment or a course of chlorambucil. Intermediate prognosis patients--ie, stage B (n = 224)--received either chlorambucil or the COP regimen (cyclophosphamide, vincristine, and prednisone). Poor prognosis patients--ie, stage C (n = 60)--received either COP or CHOP (COP plus doxorubicin). The first six courses of polychemotherapy were given every month and the other six every 3 months. 93 deaths were recorded at the time of the analysis. Stages A and B patients did not show significant differences in survival rates between treatment groups. In stage C patients 2-year survival rates were 44% with COP and 77% with CHOP (p = 0.0013, one-sided log-rank test). Disease status, as assessed by staging at 9 months, was also better with CHOP than with COP. These results led to the termination of the trial and its replacement by one in which CHOP is given to intermediate prognosis patients.

106 women between 24 and 33 weeks' gestation and in preterm labour, rigidly defined to include cervical dilatation plus regular uterine contractions, were randomly allocated to receive either intravenous ritodrine hydrochloride or no tocolytic treatment. Ritodrine treatment significantly delayed delivery for 24 hours or less but did not significantly modify the ultimate perinatal consequences of preterm labour.

Outcome of pregnancy after amniocentesis was studied in a randomised controlled trial of 4606 women, age-range 25-34 years, without known risk of genetic disease. Spontaneous abortion rate was 1.7% in the study group after amniocentesis and 0.7% in the control group after ultrasound (relative risk 2.3). In the study group, increased levels of maternal serum alpha-fetoprotein before amniocentesis, perforation of the placenta during amniocentesis, and withdrawal of discoloured amniotic fluid were associated with an increased risk of spontaneous abortion. In the first six weeks after amniocentesis/ultrasound scan, amniotic fluid leakage occurred more often in the study group but there was no difference in the rate of vaginal bleeding. Frequency of postural malformations in the infants in the two groups was the same. In the study group, respiratory distress syndrome was diagnosed more often (relative risk 2.1) and more babies were treated for pneumonia (relative risk 2.5).

450 villages in northern Sumatra were randomly assigned to either participate in a vitamin A supplementation scheme (n = 229) or serve for 1 year as a control (n = 221). 25 939 preschool children were examined at baseline and again 11 to 13 months later. Capsules containing 200 000 IU vitamin A were distributed to preschool children aged over 1 year by local volunteers 1 to 3 months after baseline enumeration and again 6 months later. Among children aged 12-71 months at baseline, mortality in control villages (75/10 231, 7.3 per 1000) was 49% greater than in those where supplements were given (53/10 919, 4.9 per 1000) (p less than 0.05). The impact of vitamin A supplementation seemed to be greater in boys than in girls. These results support earlier observations linking mild vitamin A deficiency to increased mortality and suggest that supplements given to vitamin A deficient populations may decrease mortality by as much as 34%.