OR WAIT 15 SECS
James P. Agalloco is the president of Agalloco & Associates, P.O. Box 899, Belle Mead, NJ 08502, tel. 908.874.7558, email@example.com. He is also a member of Pharmaceutical Technology’s editorial advisory board.
Understanding the purpose of the biological indicator can guide the development of an effective sterilization process.
Sterilization processes are used to ensure the safety of patients treated with products and materials expected to be sterile at time of use. The objective is to eliminate microorganisms in and on products that are introduced into the body in a manner that defeats the ordinary protections of skin, intestines, and other safeguards present. In considering patient safety with respect to sterility, a minimum requirement of one contaminated unit in a million units is considered acceptable for sterilized materials (1). The original term for this value, sterility assurance level (SAL), is non-intuitive and defining it usually entails the use of the word ‘probability’. Increasingly, this value is being called the probability of a non-sterile unit (PNSU). In routine practice, additional precautions are taken so that this minimum expectation is substantially exceeded.
The calculation of PNSU uses Equation 1, in which the lethality delivered, D-value, and initial population of the microorganism are inserted.
The equation is simple enough; however, there is a common misconception in its use. The problem lies in the incorrect use of values for population and resistance from the biological indicator rather than for the bioburden. The safety expectation relates to the routine use of a sterilizer where the bioburden is present, rather than the initial or periodic validation of the sterilization process when a biological indicator is employed. In the majority of instances, materials sterilized in conjunction with the validation exercise are not intended for patient use. The minimum PNSU as derived from the bioburden present is the critical concern.
Equation 2 estimates the PNSU for a 3-minute process at 100 °C with a starting population of 100 CFU/unit and an estimated D100 of 0.0003 minutes (2).
It should be immediately evident that this extremely short and low-temperature sterilization process provides an overwhelming margin of safety that is nearly 10,000 times greater than the minimum expectation. The moist heat resistance of the bioburden is so minimal at these conditions that there is essentially no chance for its survival (3). This is true even though the process is 3 minutes at 100 °C, not the more commonly (and wrongly expected) process performed in excess of 121 °C. The lethality of this low temperature process cannot be established with the conventional biological indicator of Geobacillus stearothermophilus, whose resistance is such that the assumed process would have no meaningful impact on its population.
Requiring destruction of a 106 population of G. stearothermophilus to the minimum PNSU expectation of 6 would require a process at 121 °C and an F0>10 minutes. Such a process offers no benefit to the patient because the bioburden will already have been killed well beyond minimum expectations at the lesser condition. If a 121 °C process delivering an F0=10 minutes were utilized instead, the PNSU would be as shown in Equation 3.
The estimated PNSU in this example would be extreme: not more than one positive in more than three million times the minimum requirement. The only justification for using such a cycle is to destroy a bioindicator that has no resemblance to the native bioburden, is present at a concentration that exceeds any reasonable real-world situation, and has extreme moist heat resistance. Killing the bioindicator is certainly safe, but this approach arbitrarily increases the adverse process impact on the product. The real target in sterilization is always the bioburden, which is generally far easier to kill. Therefore, the sterilization process should be developed with that as the objective.
The purpose of the biological indicator in sterilization is not to define the process, but rather to measure it. The steps involved in sterilization process development are outlined in Figure 1.
Figure 1. Establishing a bioburden-based sterilization process. Figure is courtesy of the author.
The activities needed to define and validate a sterilization process focused on reliable destruction of the bioburden follow a simple sequence.
Selection of a bioburden model
The resistance of the bioburden can be obtained from experimental data collected on materials prior to sterilization or based on assumptions regarding the expected bioburden. Resistance information can be obtained from the literature or experimentally determined. The United States Pharmacopeia includes a boil test that can be used to estimate microbial resistance (1,3). The boil test can be adapted to estimate bioburden D-values at the appropriate temperature if a temperature other than 121 °C is used. The population determination or estimation is straightforward.
Calculation of process duration
Inserting the population and resistance information for the assumed bioburden along with the desired minimum PNSU into Equation 1, the minimum process dwell time (F) can be determined.
Selection of the biological indicator
With the process duration established, a biological indicator with appropriate population and resistance can be identified that is appropriate for the determined process duration. The biological indicator should not be so resistant as to completely survive the process, but it should represent a meaningful challenge to confirm the required process conditions have been achieved. Partial kill of the biological indicator is most definitive as it confirms that the biological indicator possesses adequate resistance to support the process condition. Surprising as it may seem, complete destruction of the biological indicator does not provide that confirmation. Appropriate biological indicator options could include mesophilic sporeformers such as Bacillus megaterium or Bacillus oleronius (3,4).
Physical and microbiological confirmation of sterilization process
Use a combination of physical measurements and microbiological challenges to confirm that the required lethality is delivered.
Throughout this exercise, worst-case assumptions can be made to increase the confidence in the sterilization process. The typical assumptions include:
All worst-case assumptions need not be utilized, because doing so can result in a final process that is overly harsh to the quality attributes of the materials being sterilized.
There are many reasons why the bioburden should be understood as the focus of the sterilization and the bioindicator relegated to a supportive role in the validation of the process:
The validation of sterilization processes must balance the often competing considerations of increased process safety and the negative impact of over-processing. The biological indicator should be chosen to support a sterilization process that provides a reliably stable and efficacious product with an adequate margin of safety. Extending process dwell and increasing temperature merely to kill biological indicators beyond what is necessary for patient safety is never appropriate. The correct use of a biological indicator is as a measurement tool confirming sterilizing conditions have been attained within the load items sufficient to render the process sufficiently safe. Sterilization and sterility assurance need to consider bioburden destruction to safe levels as the only true objective.
1. USP, USP-NF 39, General Chapter <1229>, “Sterilization of Compendial Articles.”
2. J. Agalloco, “Increasing Patient Safety by Closing the Sterile Production Gap-Part 1-Introduction,” accepted for publication in the PDA J Pharm Sci and Tech.
3. I. Pflug, Microbiology and Engineering of Sterilization Processes (Environmental Sterilization Laboratory, Otterbein, IN, 14th ed. 2010), Table 13.7, p. 13.18.
4. M. Izumi, et al., PDA J Pharm Sci and Tech, 70 (1) 30-38 (2016).
Vol. 30, No. 4
When referring to this article, please cite it as J. Agalloco, “Kill the Bioburden, Not the Biological Indicator," BioPharm International 30 (4) 2017.