Test your cranial ultrasound interpretation skills on the web

Cranial ultrasound is now well established in neonatal units as a tool for monitoring the prevalence of intracranial lesions, timing injury, and investigating babies with abnormal neurology. Just when you thought the method was so easy that it was safe to let the trainees perform cranial ultrasound, the Hammersmith team (page 92) show that you are wrong. A welcome feature of this article is that you can privately test your own skills without fear of exposure. All the images used in this audit, and the correct answers, can be found on theArchives website (www.archdischild.com). The results of the survey were alarming, with only about a half of the answers correct. There was a low response rate from consultants, an observation open to interpretation. The authors acknowledge that reporting from a single isolated image is more difficult than from a set of images or a videotape. Nevertheless, this study has important implications for the future of clinician led cotside ultrasonography, a tradition many of us have grown up with. Should we abandon it altogether, or should we expend effort on training and improving standards of recording and reporting? An interesting subject for debate in the rapid response columns over the next weeks.

Reducing mortality in neonatal meningitis; uncertain mortality at the limits of viability

The results of the latest BPSU survey of neonatal meningitis show that the mortality fell from 20% in 1985–87 to 7% in 1996–7 (page85). There was no change in the incidence, and Group B streptococcus remains the dominant cause. Cefotaxime has replaced chloramphenicol as the antibiotic of choice, with over three quarters of the babies in this study receiving cefotaxime, often in combination with penicillin. There is a worrying rise in ampicillin resistant organisms. Whether or not these promising results will lead to a reduction or an increase in neurological morbidity in the survivors is yet to be known.

Evans and Levene (page 79), in a carefully conducted systematic review, confirm the suspicion that the population base of a preterm cohort has a significant impact on the results. Survival, particularly at 23–25 weeks of gestation, was worse in geographical cohorts than from cohorts reporting on the outcome of those who were admitted to neonatal units. These authors also highlight the range of reported outcomes, from neonatal survival, through survival to discharge to 2 year survival. Editors would be delighted if their recommendations for reporting future cohorts were followed. Perhaps in future pooling of national data within the UK and/or Europe using web based collecting systems will avoid the publication bias to which they refer.

More controversy in retinopathy of prematurity

The team in the Northern region report the outcome of a different approach to oxygen therapy in very premature babies (page 106). One hospital in their region aimed to keep babies' oxygen saturation (measured with a pulse oximeter) between 70–90% whereas the others aimed at more conventional levels of 88–98%. The unit which adopted a policy of lower oxygen saturation levels did not use arterial lines; only six of 65 survivors ever had any arterial blood gases taken. This was a controversial approach, as the authors acknowledge, breaching the BAPM guidelines and flying in the face of current standard clinical practice. Further, oxygen saturation monitoring provides no information on carbon dioxide tension or pH, both of which provide important information in achieving homeostasis. Hypocarbia is a risk factor for PVL. This was not a randomised controlled trial, which means that the findings of a lower incidence of ROP in the “low saturation” group must remain open to criticism. There was no increase in ultrasound diagnosed intracranial pathology including PVL amongst the babies monitored without arterial blood gases. Only cystic PVL is reliably detected this way, assuming that the operators had been properly trained (see also page 92). This provocative approach to early ventilator management is bound to generate heated debate.

And finally

Hazel McHaffie and her coauthors deserve credit for tackling the sensitive issue of bereavement counselling (page 125). This is something all neonatologists do, and we all think we do it pretty well, but it is salutary to be reminded of the parent's perspective of these difficult meetings. Some of their wishes could be easily met, such as an earlier appointment and one timed to coincide with a repeat visit to the obstetrician. Other requests, such as a meeting place outside the hospital, would be more difficult for most busy neonatologists to comply with. However, given the importance of these consultations we all need to consider the points made by this useful study.