Preparations from natural sources of pulmonary surfactant are subject to biological variability. Therefore Tanaka and colleagues [1] standardized their reconstituted lung surfactant by adjusting the lipid composition. The product, Surfactant TA, revealed a low variability of in vitro surface properties [2]. However, to date a convincing correlation between in vitro surface properties and physiological surfactant activity, determined as dynamic or static compliance of premature ventilated animal fetuses, has not been shown [3–5]. Nevertheless, in vitro surface properties have extensively been used for the development of artificial surfactants [6–10].

The effectiveness of several surfactant preparations for the treatment of hyaline membrane disease in premature newborns has been shown in clinical studies [1–7]. Some of these studies allow for more than one application or show a decline in the beneficial effect of replacement therapy after several hours, suggesting that at least some of the very premature babies require more than one dose of surfactant. The potential risk of substitution with exogenous surfactant may depend on the number of applications and the dose applied. Thus an optimal surfactant for replacement therapy should have high activity at a low phospholipid concentration. Artificial (chemical synthesis, cell culture synthesis) and natural (from biological sources) surfactant preparations have been extensively optimized and qualified by physical methods, such as surface tension/surface area-measurements (e. g. [8–10]). However, a convincing correlation between biochemical or physical parameters and biological surfactant activity could not be established [11, 12].