those dying in hospital with obstructed airways had less severe non-respiratory injuries than those dying with clear airways. The same tendency could be discerned when the lowest 10% were considered, although the numbers of patients were very small. These results suggest strongly that airway obstruction contributes to the death of some patients in the first 72 hours of hospital care. They do not indict airway management in those who die before they reach hospital. Discussion This study is presented as much for its method as for its results. The introduction of the ISS will do much to help to measure performance that has hitherto been the subject of speculation. Improvement in the data retrieval systems available to accident units is, however, essential if full use is to be made of auditing techniques. Because no retrieval system was available this work was based on cumulative mortality rather than the much more satisfactory concept of mortality ratio. It follows that no observations could be made on the undoubtedly important role of airway management in the survivors. The policy adopted for the prehospital care of accident victims remained constant throughout the five years under review (1971-5). Ambulance crews were trained in basic resuscitation techniques and called out hospital staff only to care for trapped patients. The results would suggest that this quality of care is optimum with regard to airway management and that no improvement in survival could be expected if more advanced training were introduced. In contrast, airway management during the first 72 hours in hospital does not appear to be satisfactory.

adrenaline concentrations, measured by specific 1. Plasma adrenaline and noradrenaline were measured in supine and upright positions in 27 essential hypertensive patients and in 12 agematched control subjects.2. Plasma adrenaline was markedly increased in a large proportion of patients, compared with control subjects, either in supine or in upright positions, whereas no significant differences were found in plasma noradrenaline.3. Six hypertensive patients were chronically treated with P-adrenoreceptor-blocking agents (five with propranolol and one with pindolol). Plasma noradrenaline was not significantly changed in the supine position but was markedly increased on standing in four out of six patients. Plasma adrenaline was decreased in five out of six patients either in supine or upright positions. 4. These findings support the view that the adrenergic system participates in the mechanism of essential hypertension.