[Prognostication of foetal weight by means of ultrasonic foetometry. Study into accuracy of various formulae and analysis of error sources (author's transl)].

The author conducted a large-scale study into all accessible formulae for foetal weight assessment on the basis of ultrasonic measurement. Included were the author's own recently devised formulae. They were all studied for their accuracy, with reference being made to 100 unselected data groupings.--Another analysis was made for the purpose of finding out the extent to which the number of data used by different examiners could change differences between estimated results derived from formulae. Reference in great detail is made also to the effects of measurement errors and selection of data upon formula results.--Proper choice of measurement level was found to be important. Abdominal parameters proved to be superior to thoracic parameters. Estimated results, generally, improved along with the number of foetal measuring points, provided they were really relevant and weightful.--Discussed is the relative value which should be ascribed to all factors with impact on the accuracy of weight prognosis. Proper choice of an adequate mathematical model or adjustment of such choice to foetal data is considered to be essential, because it has to be made by optimum consideration of all factors.--All the above formulae were verified, but accuracies claimed earlier by original authors were reproducible only in some rare cases. The demand, therefore, is made that whenever new formulae are introduced the best of all available techniques of assessment should be used to compare their results, with reference to be made to the same, unselected patients. This is considered to be the best way for an objective assessment of their advantages.--An account is given of those practicable formulae which can be recommended, as a result of the comparison, and for which easy-to-use nomograms are available.