When opiates are administered by the epidural and spinal routes, itching occurs as a side effect. We reviewed 52 reports in the literature of the use of epidural and spinal opiates to assess the incidence of itching and found an overall incidence of 8.5% in patients receiving epidural opiates, and 46% in patients receiving spinal opiates. The symptom is a recognised, though rare, side effect of systemically administered opiates, and in the case of systemic administration the itching is generalised. In the case of epidural and spinal administration, the itching may be generalised. But often a segmental distribution is demonstrable, centred on the level of injection, or the itching is localised to a particular area such as the nose and face. It is likely therefore, in the latter case, that there is an effect upon the spinal cord itself. Although occasionally spinal opiate-induced itching is extremely troublesome and lessens the value of spinal opiate pain relief, in the majority of cases, the itching is not severe and is treatable with naloxone. However, the frequent occurrence of the symptom and the likelihood of a spinal cord mechanism do provide valuable information about opioid actions, and benefit may be derived from better understanding the phenomenon. This paper states a hypothesis to explain spinal opiate-induced itch and explores the possible mechanisms of the effect.

SUMMARYIn patients with subarachnoid hemorrhage there were increased concentrations of plasma epinephrine and norepinephrine when compared with those concentrations in a group of patients admitted to hospital with other illness.Reassessment after a variable period showed that in patients whose eventual clinical result was poor the plasma epinephrine and norepinephrine concentrations increased further while in those with a good result those concentrations showed a decline. No such changes were evident in plasma dopamine-,8-hy droxylase activities which were within normal range.In a sub-group of patients who had neurosurgery after admission for clipping an aneurysm, the post-operative changes of plasma epinephrine and norepinephrine concentrations were related to the clinical condition of the patients.

SummarySubarachnoid haemorrhage from intracranial aneurysms has a poor prognosis. Operative management of intracranial aneurysms was once considered ineffective. The first 100 cases treated by microsurgery were analysed to see whether mortality and morbidity were reduced. Modern surgical techniques halved the total mortality but the morbidity was unaltered. Results can be improved by delaying surgery seven days and by treating any hypertension before surgery.

S U M M A R Y Plasma adrenaline and noradrenaline concentrations were measured in 21 patients after subarachnoid haemorrhage and in 13 control patients. Plasma noradrenaline concentrations were significantly raised in patients recovering from subarachnoid haemorrhage, confirming clinical evidence of overactivity of the sympathetic nervous system. Plasma noradrenaline concentrations in patients with a poor result were significantly higher at the time of admission than in patients with a good result, and the differences became more significant two to three days later. Therefore, the measurement of plasma noradrenaline concentrations may be a valuable test to assist clinical assessment in distinguishing between the two groups preoperatively.There is indirect evidence of autonomic nervous system dysfunction after subarachnoid haemorrhage. In the hours after a subarachnoid haemorrhage systemic hypertension is common, and electrocardiographic changes suggestive of myocardial ischaemia or even frank infarction, which have been interpreted as evidence of overactivity of the sympathetic nervous system, have been reported (Weintraub and McHenry, 1974). Greenhoot and Reichenbach (1969) reported six fatal cases of subarachnoid haemorrhage in which recent subendocardial haemorrhages and myofibrillar degeneration were evident at postmortem examination, while Smith and Tomlinson (1954) found subendocardial haemorrhage in 29 of a series of 235 patients at necropsy; such changes have been produced in experimental animals by the intravenous infusion of noradrenaline (Szakacs and Mehlman, 1960). Plasma catecholamines were measured in man after subarachnoid haemorrhage by Peerless and Griffiths in 1972 using a fluorimetric assay. They found raised plasma catecholamine levels, particularly in patients who were comatose or moribund. How-'Rhodes Scholar.

No abstract