Cytotoxic drugs for gastric and colorectal cancer.

surviving infants who gave negative results to brain stem audiometry also failed on the otoacoustic test.'4 This work also highlighted a major problem-that of validating methods of testing senses in the newborn baby. This has to be by follow up, checking the outcome with the testing methods that become possible in the older infant. Stevens's group found a poor correlation between distraction testing of the babies' hearing at 8 months of age and brain stem audiometry in the newborn, a discrepancy confirmed by others.'5 We must now be more cautious in interpreting the results of electrophysiological tests in the newborn. Babies who give negative results will need retesting several times during the first year. Though the early fitting of hearing aids is desirable, the degree of hearing impairment needs to be clearly established, particularly as maturation of the auditory pathways may be taking place, although delayed. Similar problems and challenges occur in testing vision in newborn babies.'6 1' Behavioural responses are familiar to the mother, with the baby blinking to bright light, fixating, and following a red woollen ball or a flashing light. Babies turn their heads to a diffuse light but (like turning to sound) this test may not be reliable, especially in preterm infants. All these responses give a qualitative indication of vision. Optokinetic nystagmus can be shown when a striped tape or drum is moved in a temporal to nasal direction across the newborn baby's field of vision and gives a valuable but crude indication that vision is present. Electrophysiological recording of the visual evoked potential to a flash gives limited useful information because ofgreat individual variations and because it relates as much to general cerebral function as to visual outcome.'8 Visual evoked potentials to patterns may give a measure of visual function but only after the age of 2 months. The best method of measuring visual acuity is to use the preferential looking technique. This is based on the observation of Fantz 30 years ago that patterned objects are visually interesting to infants.'9 The latest version, called the acuity card procedure, uses patterned and plain stimuli mounted in pairs on cards, and these can be used successfully even in the neonatal intensive care unit. '" Much fascinating and enjoyable research is being done into the ability of babies to discriminate and respond to smell, taste, and touch.9 We should also be glad that at long last there is widespread acceptance of the fact that newborn babies do experience real pain and need postoperative analgesia like the rest of us.20 DAVID A CURNOCK