The Quest for Thermal Death Time Validation

By Annel K. Greene, PhD, Center Director
Clemson University Animal Co-Products Research and Education Center


Validating thermal lethality of rendering processes is of importance to the industry to ensure destruction of bacterial and viral pathogens in animal by-products. The quest for a “handbook” of thermal validation has been a long-term goal of several researchers at the Clemson University Animal Co-Products Research and Education Center (ACREC). Work is on-going stepwise to reach this goal as funding has been granted. Each project has added new knowledge to the pursuit of thermal validation of rendering.

Dr. Annel K. Greene, professor and center director of ACREC, and Dr. Paul L. Dawson, professor in Clemson’s Department of Food Science and Human Nutrition, have worked on complementary studies related to bacterial safety of animal by-products. In the first research study conducted by Clemson University investigators, Greene measured the thermal conductivity of animal by-products. Thermal conductivity indicates the heat penetration per unit time into specific products.

In this study, nine different types of raw beef, pork, and poultry by-products were studied. Using the collected thermal data, Greene enlisted the help of Dawson and Dr. William C. Bridges of Clemson’s Department of Experimental Statistics. Models were created to describe the heat conductivity into each product using canning industry methodology and Food and Drug Administration-approved equipment. Due to research limitations and lack of a lab-scale rendering cooker at that time, this data was collected using non-stirred samples under retort conditions. This data added to the base of knowledge and was the first known study to measure heat penetration into animal by-products. It also served as the first introduction of Clemson University microbiologists to rendering. At this point, the learning curve began in earnest for Clemson University researchers as they learned about rendering industry processes.

The next experiment conducted in Greene’s laboratory was an exhaustive study that involved examining hundreds of poultry, pork, and beef by-product samples for microbial content. Greene enlisted the team of nearly a dozen researchers, including Dr. Zeynep Güzel-Seydim, Dr. Atif C. Seydim, and graduate student LaShanda Glenn, to help with the project. The team enumerated the raw material samples for bacterial content, then conducted thermal treatments at 260 degrees Fahrenheit (F) for 40, 60, 80, and 100 minutes and counted residual populations.

The results of this long-term study were wholly erratic and confusing. After testing multiple sub-hypotheses to understand why bacterial populations seemed to decrease in one sample only to increase in the next sample over treatment time, the researchers eventually realized that the classical standard techniques used for bacterial enumeration of all food products resulted in completely erroneous bacterial counts when used on high-fat containing animal by-products. Therefore, the results of the enumeration part of the study were in question and researchers realized they needed to develop new methodology before they could proceed.

A second goal of this study was to identify microorganisms that may survive thermal processing at 260 degrees F for 40, 60, 80, and 100 minutes. Identification of isolated microorganisms was performed using gas chromatograph fatty acid methyl ester analysis using software from MIDI, Inc., Newark, DE. Microorganisms isolated from the laboratory-cooked poultry by-products included Staphylococcus aureus, Staphylococcus warneri, Staphylococcus epidermidis, Bacillus subtilis, Bacillus licheniformis, Bacillus pumilus, Bacillus lentimorbus, Bacillus marinus, Escherichia coli, and Klebsiella trevisanii.

A third goal of the study conducted by Dawson and his laboratory was to determine the thermal inactivation profile of heat resistant bacteria isolated from raw animal by-products. Due to the nature of the raw material, a wide range of bacterial species can be present. Numerous studies have indicated that heat resistant bacteria can develop from repeated exposure to high temperatures. Dawson, research associate Dr. Inyee Han, and graduate student Lina Ramirez-Lopez obtained bacteria from raw poultry offal. Using laboratory conditions and a simulated matrix, they applied a stepwise increase in heat intensity until only a few bacteria were recovered. These cells were selected and further characterized for heat resistance and other traits. The heat resistant isolates were found to be spore formers that grew under aerobic conditions. However, interestingly, these organisms did not display linear heat inactivation properties that are typically characteristic of spore-forming bacteria. Working with Bridges, Dawson developed three statistical models to reflect the heat inactivation profiles that better described the microbial characteristics than the traditional linear model. Using low-end treatment temperatures of 196 degrees F, 201 degrees F, and 205 degrees F, the populations of bacteria demonstrated a downward stair-step pattern during heat treatment. Statistical models were developed to predict the heat inactivation rate of the isolate and to help researchers better understand how heat resistant bacteria isolated from raw animal by-products act in simulated by-products.

The next project involved trying to map the interior of a continuous rendering cooker. The challenge was to validate that there are no cold spots within a cooker so that the minimum amount of time the product was treated at a particular temperature could be measured. However, due to the physical nature of the rendering cooker and the inability to obtain telemetry data out of a cooker, this direct approach could not be accomplished.

Therefore, after much deliberation, Greene suggested using a “holding tube” concept similar to the dairy industry where time and temperature are measured after the heating process. It was recommended that the cooker drainer assembly already installed on most cookers could serve as the “holding tube.” With assistance from Rob Horton of The Dupps Company, the “holding time” during which temperature could be determined post-cooker was measured. It is believed that this post-cooker time may provide sufficient thermal processing to validate thermal lethality and this information is being used to build additional knowledge in current thermal death time studies.

In this same time frame, a project was initiated to determine the thermal death time of avian influenza in animal by-products. The concerns about avian influenza impacting the industry were increasing at this time due to outbreaks in Asia. Dr. Thomas R. Scott, professor of Animal and Veterinary Sciences at Clemson University and now dean of the College of Agriculture, Forestry, and Life Sciences, spent many months developing an enzyme-linked immunosorbent assay (ELISA) for testing avian influenza virus proteins in animal by-products. However, testing the ELISA method with the purified virus proteins in rendered poultry products revealed that the high fat matrix interfered with the binding of the antibodies and resulted in a false test. Once again, the unique nature of animal by-products challenged the researchers. Therefore, a polymerase chain reaction assay was chosen to detect Type A avian influenza ribonucleic acid (RNA).

Greene, Scott, Dr. Adam Leaphart of the Clemson University Livestock Poultry Health/Veterinary Diagnostic Center, and laboratory technician Laine Chambers conducted thermal trials using animal by-products spiked with an inactivated strain of highly pathogenic avian influenza virus (A/turkey/Wisconsin/68 H5N9), which was purchased from the U.S. Department of Agriculture National Veterinary Services Laboratory in Ames, IA. This material provided the researchers with a non-viable strain of highly pathogenic avian influenza with an intact viral capsid that could be used under biosafety level two conditions. After thermal trials, analysis of data with the assistance of Bridges revealed that Type A influenza RNA was destroyed in 30 seconds or less at any temperature between 230 degrees F to 284 degrees F. This temperature and time is well within the “holding time” measured on commercial cookers in previous experiments.

It is known that some microorganisms such as spore formers are more heat resistant than non-spore formers and viruses. Therefore, it is possible to surmise that the temperature and time conditions that would be sufficient to kill an extremely heat resistant organism would be more than sufficient to kill a less heat resistant organism in the same matrix. This approach is currently being pursued by Clemson researchers.

To that end, work is underway at Clemson University to determine the D values for various microorganisms in animal by-products at rendering cooking temperatures. The D value is the time required to inactivate 90 percent of a microbial population at a specific temperature in a specific medium. The new bacterial enumeration methods developed by Clemson University for use with high-fat containing animal by-products are being employed. ACREC researchers have embarked on a multi-year study conducting thermal trials on animal by-products.

Greene and her current graduate student Yubo Zhang conducted research on thermal death time analysis of animal by-products using Geobacillus stearothermophilus as a test organism. The purpose of this study was to determine the thermal lethality that may be applied by rendering. G. stearothermophilus is a spore-forming bacterium that is often used as a surrogate for thermal processing studies in the food industry because it is capable of withstanding extreme heat.

G. stearothermophilus is reported to be up to 20 times more heat resistant than the heat stable Clostridium botulinum. Therefore, inactivation of G. stearothermophilus would indicate thermal lethality exists for the destruction of other microorganisms and especially destruction of vegetative (non-spore) organisms. The purpose for using this organism first in these trials was because it is non-pathogenic, which simplifies laboratory safety requirements. Additionally, it was hypothesized that if this extraordinarily heat resistant microorganism could be killed under rendering conditions, then less heat resistant organisms such as Salmonella and Clostridium perfringens should be killed.

G. stearothermophilus spores were added to poultry by-products at the rate of approximately one million spores per gram of material. Using the upper end of the typical rendering temperature range, Zhang measured an approximately five log reduction in G. stearothermophilus spores caused simply by the come-up time (three to four minutes), which was required to reach the treatment temperature. During the study, additional failures in methodology were determined and again, the team had to invent additional bacteriological methods to allow data collection.

Work is continuing on determining thermal destruction across a variety of temperature ranges for treatment of G. stearothermophilus spores in animal by-products. Destruction of the extremely heat resistant G. stearothermophilus spores under rendering conditions would be excellent data to indicate that other, less heat resistant organisms could be destroyed under similar conditions.

Upon completion of the G. stearothermophilus trials, the researchers want to continue testing thermal death time of the pathogens Clostridium perfringens and a cocktail of four_ Salmonella_ species of concern to rendering.

Data collection studies continue at Clemson University in order to derive this information for the rendering industry with the ultimate goal to develop a “handbook” of D values for microorganisms of concern in animal by-products.


October 2011 RENDER | back