No abstract

Summary Treatment of the pregnant woman with melanoma is contentious. With the aid of a computerized melanoma register, in which pregnancy data can be recorded, 290 women with melanoma were reviewed (249 alive and 41 dead). Overall, 23 patients were pregnant at ithe time of diagnosis of melanoma, and another 23 became pregnant at some time after primary treatment of a melanoma. Pregnancy appears to have no significant influence on the survival of patients with melanoma J but it is recommended that pregnancy should be avoided for the first three years following excision of a melanoma.

Treatment of the pregnant woman with melanoma is contentious. With the aid of a computerized melanoma register, in which pregnancy data can be recorded, 290 women with melanoma were reviewed (249 alive and 41 dead). Overall, 23 patients were pregnant at the time of diagnosis of melanoma, and another 23 became pregnant at some time after primary treatment of a melanoma. Pregnancy appears to have no significant influence on the survival of patients with melanoma;! but it is recommended that pregnancy should be avoided for the first three years following excision of a melanoma.The management of pregnant women with a melanoma can be difficult. The literature is confusing, and reports on the behaviour of melanoma in pregnancy are often contradictory. A study of a melanoma register was therefore undertaken to try to clarify this situation by asking three questions: (1) Is there a difference between pregnant and non-pregnant women in the occurrence of melanoma at various sites? (2) Do women who develop melanoma while pregnant have a worse prognosis than non-pregnant women? (3) Does a subsequent pregnancy have any effect on the prognosis of melanoma? Patients and methodsA melanoma register has been in use at Frenchay Hospital since the early 1950s. This was originally maintained as a card index but, in 1980, the records from 1967 onwards were transferred to a computer (Briggs et al. 1984) with an Frenchay Hospital, Bristol BS16 1LE Department of Reconstructive Surgery D. S. McMANAMNY Registrar in Plastic Surgery

To establish the stages of brain processing in an auditory stimulus localization task, event‐related potentials (ERPs) were recorded from 24 normal subjects listening to brief white noise stimuli in a free‐field situation from front, back, left and right loudspeakers. The subject's task was to respond to ‘target’ stimuli from one designated speaker. Performance varied as a function of sound location, stimuli in the front/back dimension being more difficult to localize than those in the left/right. ERP results, based on averaged waveforms, difference waveforms and the factors derived from a principal components analysis, revealed a series of task related components. Some were relatively transient, others more sustained in character. One brain component showed task related amplitude changes with an onset as early as 15 msec. These changes were target specific for sounds in the easier left/right dimension only. Later components, such as the P300, also varied in amplitude between targets and non‐targets, but showed no significant amplitude or latency differences to target stimuli as a function of location, despite the significant performance differences.

Late components of brain event-related potentials reflect aspects of selective attention, stimulus evaluation, and possibly memory update mechanisms. Several of these components were measured during an auditory target detection task, performed by 20 schizophrenic and 20 normal subjects. Both the amplitude of those components and a more general late amplitude measure were significantly reduced in schizophrenics, for both target and non-target stimuli. One general late amplitude measure, from the scalp vertex, could alone correctly classify 85% of patients and 95% of controls. The source of these differences may lie in a protracted positive potential shift.