Abstract
In the production of large-volume parenterals in Japan, equipment and devices such as tanks, pipework, and filters used in production processes are exhaustively cleaned and sterilized, and the cleanliness of water for injection, drug materials, packaging materials, and manufacturing areas is well controlled. In this environment, the bioburden is relatively low, and less heat resistant compared with microorganisms frequently used as biological indicators such as Geobacillus stearothermophilus (ATCC 7953) and Bacillus subtilis 5230 (ATCC 35021). Consequently, the majority of large-volume parenteral solutions in Japan are manufactured under low-heat sterilization conditions of F0 <2 min, so that loss of clarity of solutions and formation of degradation products of constituents are minimized. Bacillus oleronius (ATCC 700005) is listed as a biological indicator in “Guidance on the Manufacture of Sterile Pharmaceutical Products Produced by Terminal Sterilization” (guidance in Japan, issued in 2012). In this study, we investigated whether B. oleronius is an appropriate biological indicator of the efficacy of low–heat, moist-heat sterilization of large-volume parenterals. Specifically, we investigated the spore-forming ability of this microorganism in various cultivation media and measured the D-values and z-values as parameters of heat resistance. The D-values and z-values changed depending on the constituents of large-volume parenteral products. Also, the spores from B. oleronius showed a moist-heat resistance that was similar to or greater than many of the spore-forming organisms isolated from Japanese parenteral manufacturing processes. Taken together, these results indicate that B. oleronius is suitable as a biological indicator for sterility assurance of large-volume parenteral solutions subjected to low-heat, moist-heat terminal sterilization.
- Biological indicator
- Bacillus oleronius
- Bioburden
- Terminal sterilization
- Moist-heat sterilization
- Large-volume parenteral solution
Introduction
Many large-volume parenteral products for clinical use are manufactured on a large scale in Japan, and these products are available in various concentrations and volumes to meet various clinical conditions and are used as they are without further admixing and compounding procedures, particularly helping reduce the time and labor spent by healthcare staff on tasks including such admixing and compounding before administration. Typically, these products are distributed in plastic containers equipped with a thick and rugged rubber stopper that contains a port suitable for insertion of a plastic or stainless steel needle, both of which are currently used in Japan. The design and labeling of these containers are extremely important from the viewpoint of reducing potential medical and medication errors (1). With regard to sterilization, the majority of large-volume parenteral solutions manufactured in Japan are sterilized under low-heat, moist-heat sterilization conditions of F0 <2 min (Table I). The main reason for employing low-heat sterilization conditions is the empirical consensus that the solutions have to be “clear in clarity and colorless” as far as possible (except for fat emulsions or coloration due to addition of vitamins or trace elements), because physical changes such as incompatibility of parenteral solutions when admixed with additive medications can be immediately visually detectable. Another reason is to minimize formation of 5-hydroxymethylfurfural (5-HMF), a degradation product of glucose formed during heat sterilization (2), because large-volume parenteral solutions generally contain glucose.
Terminal Sterilization Conditions and Amount (Units) of Large-Volume Parenteral Solutions Manufactured from 2005 to 2010 in Japan
Of course, highly transparent containers are also required to enable large-volume parenteral solutions to be visually inspected for physical changes such as discoloration of the solution. Interestingly, polyvinyl chloride (PVC) was not generally used as a packaging material for such products in Japan even before the problem of plasticizer leaching was identified (3), because there were general opinions at the time that PVC bags, though transparent enough, do not permit adequate visual inspection of the liquid surface of the solution during administration, that is, it is somewhat difficult to confirm the liquid surface to estimate the remaining volume during administration.
Sterilization conditions employed in the manufacture of large-volume parenteral solutions in other countries are typically more severe (e.g., 121 °C for 15 min or more) than those employed in Japan, where sterilization conditions of low F0-value are considered acceptable, because the equipment and devices such as tanks, pipework, and filters used in the production processes are exhaustively cleaned and sterilized, and the cleanliness of water for injection, drug materials, packaging materials, and manufacturing areas is well controlled, so that the bioburden is low. In a recent survey (Table II), it was found that the bioburden of large-volume parenteral solutions manufactured in adequately controlled environments is less heat resistant than Geobacillus stearothermophilus or Bacillus subtilis 5230 (ATCC 35021) used in sterilization validation of terminally sterilized products. Table II shows details of the bioburden identified in our recent survey. The results of the survey are consistent with another bioburden survey (including environments) conducted by PDA Japan Chapter, which found that D-values of spore-forming microorganisms, including those detected in production environments, are not more than 0.05 min at 121 °C (4). As such, bioburden population and low bioburden resistance are also one of the essential factors supporting the low-heat stabilization processes.
Results of Survey on Heat-Resistant Spores Identified in Production Environments of Japanese Manufacturers of Large-Volume Parenteral Solutions
Usually, manufacturers of large-volume parenteral solutions perform challenge tests with biological indicators to validate sterilization, using the most heat-resistant microorganism isolated from their own manufacturing environments. However, in order to increase the reliability and uniformity of sterilization validation, it might be better to perform challenge tests with standardized biological indicators purchased from recognized culture collections instead of in-house biological indicators. Therefore, candidate biological indicators were selected from a stock of microorganisms registered in cell banks such as ATCC, and their suitability for sterilization validation was explored with the support of the Intravenous Solutions Society (an association of manufacturers of large-volume parenteral solutions in Japan). As a result, Bacillus oleronius (ATCC 700005) was picked out as a candidate indicator for common use, because this microorganism showed the greatest heat resistance (D121 = 0.03 min and D105 = 1 min) among the microorganisms identified in environmental surveys conducted by member companies of the Intravenous Solutions Society. It should be noted that this microorganism is already included in the “Guidance on the Manufacture of Sterile Pharmaceutical Products Produced by Terminal Sterilization” (guidance in Japan, issued in 2012) (5). In the present study, the working group established in the Intravenous Solutions Society (Table III) performed a detailed examination of B. oleronius in order to confirm its suitability as a biological indicator for sterilization validation of large-volume parenteral solutions. Here, we report the results of the study, including characterization of this microorganism and establishment of test methods for D-values.
Working Group for Biological Indicators (Intravenous Solutions Society*)
Materials and Methods
Study on Bacillus oleronius
Initially, testing conditions such as culture media, incubation conditions, and heat-resistance parameters were optimized and uniform protocols were established to ensure consistency of testing conditions among member companies of the Intravenous Solutions Society. Then, the optimized tests were used to examine the suitability of B. oleronius as a biological indicator for sterilization validation. B. oleronius is hereafter referred to as Test BI. The results of optimization tests are included in the Materials and Methods section. All determinations of D-value and z-value were conducted in accordance with International Organization for Standardization (ISO) 11138, and the specific method was the survivor curve method.
1. Selection of Culture Media for Spore Formation and Purification of Spores:
Each member company purchased freeze-dried Test BI from ATCC and incubated it in SCDA (soybean casein digest agar) medium according to the package insert instructions to obtain the test microorganism. The test microorganism was incubated in four kinds of media (6), SCDA, standard agar medium, TYEA (tryptose yeast extract agar), and YNA (yeast extract nutrient agar), at 30–35 °C for 7 days. Finally, metallic salt-containing YNA medium was adopted as the spore-forming medium (Figure 1), because it provided a consistent spore formation rate of more than 80%.
Culture of Test BI (Bacillus oleronius) in YNA medium on day 7 of cultivation (left) and spore formation (right).
Formula of YNA with metallic salt solution
2. Heating Apparatus and Test Containers for Heat-Resistance Tests:
A survey of member companies of the Intravenous Solutions Society showed that the heating apparatus routinely used is an oil bath or a biological indicator evaluation resistometer (BIER), with a glass ampule or a capillary as a test container. Thus, D-values of Test BI (specifically, spores cultured in spore-forming media) in water for injection were measured using the heating apparatus and containers routinely used by member companies. The D-values obtained were essentially equivalent (Tables IV and V). Thus, all member companies performed subsequent tests under the following conditions:
Heating apparatus: Use a BIER or an oil bath.
Test container: Use a glass ampule or a glass capillary.
If a glass ampule is used, its volume should be 1 mL or less, and the volume of the test solution should be 0.1 mL.
Comparison of D-Values of Test BI in Water for Injection Measured in Different Heating Apparatus (BIER and Oil Bath)
Comparison of D-Values of Test BI in Water for Injection Measured in Different Test Containers (Glass Capillary and Glass Ampule in Oil Bath)
3. D-Values and z-Values of Test BI in Large-Volume Parenteral Solutions:
Generally, large-volume parenteral products can be categorized by their intended usage (provision of water, provision of cellular fluids including electrolytes, and provision of nutrients), and their formulations differ greatly by product category. Thus, D-values were first measured according to category of products (Table VI) and then, based on these results, products were selected for measurement of z-values (Table VII).
D-Values Measured in Various Large-Volume Parenteral Solutions
z-Values Measured in Various Large-Volume Parenteral Solutions
4. Measurement of D-Values of Test BI in Large-Volume Parenteral Solutions with Different Glucose Concentrations and in Phosphate Buffer Solutions with Different pH Values:
Among microorganisms commonly used as biological indicators, the D-values can change with glucose concentration (7). Thus, D-values of Test BI were measured at increasing glucose concentrations (Table VIII).
D-Values of Test BI (Measured in Glucose Solutions with Different Glucose Concentrations)
Also, since spore survival of heat-resistant microorganisms exposed to heat treatment is reported to change with the amount of phosphate buffer solution added to parenteral solutions (8), D-values in this study were measured using phosphate buffer solutions of three different pH values (Table IX). The D-values were greatly decreased at pH 3.5. To investigate the reason for this, bacterial counts were measured after suspension of the spores and immediate transfer to SCDA medium, and also after suspension of the spores, followed by a standing period of 1 h, and then transfer to SCDA medium. The bacterial counts are shown in Table X.
D-Values of Test BI (Measured in Buffer Solutions of Different pH Values)
Changes in Bacterial Counts in a Buffer Solution of pH 3.5
5. D-Values of Test BI Measured in Stored Spore Suspensions:
To investigate the stability of Test BI during storage, spores suspended in water for injection were stored in a refrigerator (8 °C or below) for 1 year and the D-values were measured (Table XI).
D-Values of Test BI Measured in Water for Injection Using Spore Suspensions Stored for More than 1 Year
6. Spore-Forming Ability after Repeated Passages in Culture (Up to 20 Passages) and Measurement of D-Values of the Spores:
To investigate alterations or mutation of Test BI during passaging, 20 passages were performed (Figure 2). The D-values were measured at each passage using water for injection (Table XII).
Colonies of Test BI (Bacillus oleronius) after serial passage (left), and microscopic images (right).
D105-Values of Test BI in Spore Suspensions after Various Numbers of Passages in Culture
Discussion and Conclusions
In Japan, large-volume parenteral solutions are manufactured in well-controlled production environments in which the target bioburden of products before sterilization is set as low as <1 colony-forming unit (CFU)/unit, and this is achieved by various means, including the use of the ISO Class 5 (not Grade A) environment in filling and sealing areas. Control of the bioburden enables products to be sterilized under low-heat conditions to minimize chemical and physical changes, such as formation of glucose degradation products (GDPs) and discoloration of solutions. Although GDPs have long been considered safe, recent reports indicate that they combine with serum proteins to form carboxymethyllysine, which may interfere with cellular activity and impair cardiac and renal functions (9). The amounts of 5-HMF and GDPs produced in peritoneal dialysis solutions during moist-heat sterilization are thus a potential safety concern (10). On the other hand, the count of viable microorganisms was less than 5 CFU per 25 mL when a product solution was filled in a vial through a 0.22 μm filter and then received heat treatment at 90 °C for 15 min, while no viable microorganisms remained when the vial was further conducted with heat treatment at 100 °C for 10 min (11). Therefore, careful biocontrol of production environments is extremely important to enable the use of low-heat conditions that provide adequate sterilization while minimizing formation of degradation products in parenteral solutions.
An important requirement in low-heat sterilization is reliable validation of the sterilization conditions, using challenge tests with appropriate biological indicators. For this purpose, common biological indicators may be better than so-called in-house biological indicators to ensure uniformity among manufacturers. Here, we focused on B. oleronius (ATCC 700005) as a candidate common indicator, because this microorganism exhibited a D-value similar to those of the most resistant microorganisms identified in the environmental survey conducted by member companies of the Intravenous Solutions Society, and also because it showed the highest heat resistance (D121 = 0.03 min and D105 = 1 min) compared with various other candidate microorganisms.
Therefore, we set out to evaluate the sterilization characteristics of B. oleronius. It is well established that the D-values of microorganisms in large-volume parenteral products vary according to the constituents, which may include inorganic substances, carbohydrates, and amino acids (7). This was also the case for D-values and z-values in the present study (Tables VI⇑–VIII). We found that D-values of B. oleronius increased with an increase of glucose concentration (Table VIII), in accordance with findings in other microorganisms (7). This increase is considered to be a result of dehydration of the spores due to the change of osmotic pressure (12–14). The effect of pH was also examined in phosphate buffer solutions. The D-value was significantly decreased at low pH (pH 3.5) (Table IX). However, there was little change in the number of microorganisms up to 3 h (Table X). Therefore, we found that exposure to low pH value solution did not influence extinction of Test BI, and the D105 value of 0.36 was precisely measured with appropriate test methods and conditions. These results are consistent with a previous report that treatment of spores with citrate buffer of pH 3.5 releases surface-bound metals such as Ca into the supernatant, increasing the thermolability of the spores, and thus causing a decrease of the D-value (15). In addition, the D-value showed no change when spores suspended in water for injection were refrigerated for 1 year (Table XI). We also confirmed that B. oleronius cultured up to the 20th passage exhibited no change in D-values (Figure 2 and Table XII). All of these observations indicate that this microorganism has excellent characteristics as a biological indicator for large-volume parenteral solutions. This view is consistent with the fact that B. oleronius (ATCC 700005) is already listed as a biological indicator in the guidance on the manufacture of sterile pharmaceutical products produced by terminal sterilization (5), and its sterilization resistance is well established.
In conclusion, there are considerable advantages to the manufacture of large-volume parenteral solutions with sterilization conditions of low F0 values in terms of both product quality and manufacturing cost (16), but it is important to ensure that safety is not compromised. Needless to say, bioburden testing of each lot of product should be a requirement with the use of this Test BI. Additionally, the heat resistance of any spores detected should be characterized and compared with the resistance of Test BI, which will support that the validation with this Test BI is ongoing. We believe that the reliability of sterility assurance in the production of parenteral products can be further improved by using a common biological indicator to validate low F0 sterilization conditions. Here, we focused on B. oleronius as a candidate indicator. The results of the present study confirmed that its characteristics are appropriate for this purpose. Thus, we believe adoption of B. oleronius as a common indicator will enhance patient safety by improving the reliability and uniformity of low F0 sterilization conditions during the manufacture of large-volume parenteral solutions.
Conflict of Interest Declaration
The authors declare no conflict of interest.
Acknowledgments
We thank Dr. Tsuguo Sasaki and Dr. Tomihiko Koshikawa for providing helpful advice and insight regarding pharmaceutical raw materials, production environments, sterilization assurance, and test methods for spores of B. oleronius.
- © PDA, Inc. 2016
