journal article
Download Only Collection
Home
doi: 10.1136/bmj.303.6800.435pmid: 1912833
OBJECTIVE--To investigate the incidence of fatal myocardial infarction in women in the two randomised arms of the Scottish adjuvant tamoxifen trial. DESIGN--Retrospective review of hospital notes to determine with the greatest possible certainty women who had died of an acute myocardial infarction. SETTING--Scottish Cancer Trials Office, the University of Edinburgh. PATIENTS--1070 postmenopausal women with operable breast cancer who were randomised to receive either adjuvant tamoxifen for five years or until relapse (539 patients) or tamoxifen for at least six weeks on the confirmation of first recurrence (531 patients). MAIN OUTCOME MEASURES--Incidence of fatal myocardial infarction in women with no known or suspected systemic cancer. RESULTS--Of the 200 women who died in the adjuvant tamoxifen arm of the trial, 44 were free of cancer at death and 10 of these died of myocardial infarction. In the observation arm 251 women died, of whom 61 showed no evidence of systemic cancer and 25 had a fatal myocardial infarction. The incidence of fatal myocardial infarction in the two groups was significantly different (chi 2 = 6.88, p = 0.0087). CONCLUSION--Tamoxifen given for at least five years as adjuvant therapy for breast cancer seems to have a cardioprotective oestrogen-like effect in postmenopausal women.
Gulliford, M C; Petruckevitch, A; Burney, P G
doi: 10.1136/bmj.303.6800.437pmid: 1912834
OBJECTIVE--To determine whether length of delay before treatment; specialty and grade of the surgeon; and use made of surgery, radiotherapy, and chemotherapy influenced the survival of patients with cancer of the bladder, after adjusting for case severity. DESIGN--Retrospective cohort study. SETTING--South East and South West Thames health regions. PATIENTS--609 men aged under 75 resident in the South Thames regions who had been registered as new cases of bladder cancer in 1982, 35 of whom were excluded, leaving 574 eligible patients. Analysis was based on 75% retrieval rate for case notes. MAIN OUTCOME MEASURES--Duration of survival from date of diagnosis of the bladder tumour. RESULTS--10 prognostic variables were used to adjust for case severity. The median delay from referral to first treatment was 48 (interquartile range 27-84) days. Treatment after a short delay was associated with shorter survival because of the early treatment of more severe cases. Consultants treated 68% of patients, trainee surgeons treated less severe cases. Initial treatment was by a urologist in 67% of cases, but the specialty of the surgeon was not associated with prognosis. The associations of radiotherapy, cystectomy, and systemic chemotherapy with survival were interpreted in terms of selection bias as well as therapeutic effect. CONCLUSION--Case severity was the most important influence on survival and influenced length of delay before treatment, grade and specialty of the surgeon, and main treatment allocation. After adjusting for case severity variations in these processes of care were not strongly associated with variations in survival.
Howard, A J; Dunkin, K T; Musser, J M; Palmer, S R
doi: 10.1136/bmj.303.6800.441pmid: 1912835
OBJECTIVE--To investigate the epidemiology of invasive disease due to Haemophilus influenzae type b, the clones responsible, and the antibiotic resistance of the isolates. DESIGN--Prospective population based analysis of clinical and epidemiological data collected for Gwynedd during 1980-90 and in the whole of Wales during 1988-90. SETTING--19 hospitals in Wales; all medical microbiology laboratories in Wales participated. PATIENTS--82 patients with confirmed invasive infections caused by H influenzae type b in Gwynedd during 1980-90 and 207 in Wales during 1988-90. MAIN OUTCOME MEASURES--Clinical and epidemiological measures; analysis of the clonal types of the isolates based on the electrophoretic mobilities of 17 metabolic enzymes; and antibiotic resistance. RESULTS--The annual incidence of H influenzae type b infections in Gwynedd was 3.2 cases/100,000 and in Wales was 2.5 cases/100,000. Most cases occurred in children aged under 5 years, the highest annual incidence being in those aged under 1 (84.6/100,000 and 56.9/100,000 in Wales). The cumulative risk of acquiring H influenzae type b disease by the fifth birthday was one in 456 in Gwynedd and one in 578 in Wales. Fifteen per cent of cases in Gwynedd and 7% of those in Wales occurred in adults. Predominant clinical conditions were meningitis in children and pneumonia in adults. In Gwynedd 2/70 (3%) children and 5/12 (42%) adults died. Long term neurological sequelae occurred in 8% (4/48) of children who survived haemophilus meningitis. Children presenting with infection were usually the youngest members of their family. No secondary household cases were identified. 100 of 128 (78%) strains were of a single clone, electrophoretic type 12.5, and 4/207 (1.9%) isolates from Wales were resistant to both ampicillin and chloramphenicol. CONCLUSIONS--The annual rate of infection in children aged under 5 in four Welsh counties was 12-44% higher than that previously published for the United Kingdom. The study emphasises the potential value of a vaccine effective in early infancy and provides baseline data to assess its efficacy after its introduction. Alternatives to ampicillin and chloramphenicol should be used as first line, empirical treatment for severe infections that might be caused by H influenzae type b in Wales.
Showing 1 to 10 of 48 Articles