|Home | About | Journals | Submit | Contact Us | Français|
It was heartening to read an article by a senior obstetrician—Mary Macintosh—describing the difficulty experienced by the Confidential Enquiry into Stillbirths and Deaths in Infancy when assessing the interpretation of fetal heart rate monitoring patterns during labour (January 2001, JRSM, pp. 14-16). Like her, I have experienced difficulty in drawing robust causal inferences from observational studies1, and I agree with her that the existing randomized controlled trials (RCTs) can not rule out the possibility that mortality is lower with continuous fetal heart rate monitoring than with intermittent auscultation. Data from the Cochrane systematic review2, for example, are compatible with a 40% reduction in the odds of intrapartum and first week deaths of normally formed babies—a reduction that many people would consider important even though such deaths are now very rare indeed.
What prospects are there for obtaining unbiased and more precise estimates of the effects of continuous monitoring on mortality and, perhaps, cerebral palsy? Sample size certainly presents a challenge, but this should not be regarded as insuperable. More than a decade ago, a multinational RCT quickly showed, after cluster randomizing 70 000 women, that routine formal fetal movement counting is unlikely to be a very effective way of reducing fetal deaths3. Such multicentre studies not only make large studies easier to mount, they also provide evidence that may be more widely applicable. The recently published trial of elective caesarean section for breech presentation at term exemplifies this impressively4.
Pending decisions about whether further randomized trials are feasible, the existing evidence2 can still inform practice. The reduction in neonatal seizures associated with continuous fetal heart rate monitoring has only been seen in controlled trials in which this screening test was used in conjunction with an assessment of fetal acid—base status to rule out ‘false positives’2. There is no robust evidence to support the use of continuous fetal heart rate monitoring alone, which simply increases the use of caesarean section, with no evidence of any compensating beneficial effects.
Just as long as the evidence Mary Macintosh has reviewed remains so slim, it will be difficult to know what should constitute ‘quality’ in the interpretation of fetal heart rate traces. She mentions that ‘a national evidence-based guideline funded by the Department of Health is in preparation’. I hope that it will be recognized that the most rational starting place for developing guidelines for interpreting fetal heart rate traces is the practice within those controlled trials, such as the Dublin study5, which have shown a beneficial effect on neonatal seizures.