[Peculiarities of respiratory mechanical studies in spontaneously breathing newborn infants. 2: Evaluation technics, parameter distribution and diagnostic validity].

Manual evaluation techniques are commonly used until now for the evaluation of recordings. They will be replaced by a computer-aided evaluation of the measuring signals which has several advantages, e.g. reduction of manual expense, processing of high quantities of data, application of powerful algorithms for computation. Both methods are discussed shortly. Hitherto a lot of reference values for the breathing mechanics in newborns have been set up but the various parameters show a wide scatter. This reduces the diagnostical value of breathing mechanical investigations and the clinical interpretation of the measuring results is rendered more difficult. Possibilities for the reduction of the scatter of parameters consist in standardizations of the measurement technique, the investigative conditions and the evaluation techniques as well as in the evaluation of a higher number of breaths in order to get characteristical values for the actual clinical state. The judgment of the diagnostical value of breathing mechanical investigations is commonly based on the subjective assessment by the investigator. Until now systematic investigations for the quantification of the diagnostic value (sensitivity, specifity, predictive value, ...) are not known, because the computation needs a high quantity of data from healthy newborns and such with respiratory distress. The number of cases in the different publications is often too small for these computations. The missing standardization makes it for some purposes impossible to combine the results of the different working groups.