[Quantitative description of the resistance of microbiological indicators by means of characteristic data (author's transl)].

The ambiguity of the quantitative description of the resistance of microbiological indicators by means of the terms common thus far, namely "resistance" (maximal period of action after which all indicators still exhibit organisms capable of reproduction) and "death time" (minimal period of action after which organisms capable of reproduction cannot be demonstrated any more in any of the indicators) is shown. Among other factors, the probability that all indicators will yield identical findings (sterility or growth of surviving organisms) at a defined time, is dependent upon the number of indicators in each case. The more indicators are tested, the lower "resistance" and the later the "death time" will be. Instead of times when all indicators present an identical reaction, times (or values related to these times in a defined way) should be taken at which indicators show the absence of organisms or the presence of surviving ones with a defined frequency. There is a defined relationship between the frequency of indicators found free from (resp. containing) surviving organisms and the number of surviving organisms per indicator (cf. Figs. 1 and 2). If the regularity of the decrease of organisms with the period of action of the destroying agent is known, 2 values are sufficient for a clear description of the resistance of a preparation of microbiological indicators. As characteristic values those values of action are proposed after which 99% and 1% of indicators will exhibit surviving organisms (t99% and t1%, respectively). The period during which the dependence of the frequency of presence of sterile indicators (po) upon the period of action of the destroying agent may be determined with sufficient precision and at justifiable experimental expenditure, is corresponding to a decrease of surviving organisms to appr. 1/30. From practical experience, it will be justified in the preponderant majority of cases to assume a logarithmic orderof death as given. Under such conditions, there will be a clear relationship between lg(-ln po) and the period of action of the destroying agent, so that test results may be easily evaluated by means of a simple graphical method or regression calculation. Graphical (FIG. 3) and mathematical determination of characteristic values from the frequency of indicators exhibitoring surviving organisms in the destruction test, after different periods of action, is explained by means of an example (Fig. 3; Tables 2 and 3) and a calculation (Table 1). To keep variation of experimental results sufficiently low, at least 30 (if possible 50) indicators per period of action should be studied for surviving organisms (cf. Fig 4). When evaluating results, different weights of the individual values and their asymmetric distribution around a maximum at t80% (Fig 5) should be taken into consideration.