Redbook 2000: IV.C.4.a Subchronic Toxicity Studies with Rodents
Toxicological Principles for the Safety Assessment of Food Ingredients
Chapter IV.C.4.a. Subchronic Toxicity Studies with Rodents
Return to Redbook 2000 table of contents
- Good Laboratory Practice
- Test Animals
- Test Substance
- Experimental Design
- Observations and Clinical Tests
- Necropsy and Microscopic Examination
Subchronic toxicity studies with rodents are generally conducted for 90 days (3 months), but they may be conducted for up to 12 months. Results of these studies (1) can help predict appropriate doses of the test substance for future chronic toxicity studies, (2) can be used to determine NOELs for some toxicology endpoints, and (3) allow future long-term toxicity studies in rodents and non-rodents to be designed with special emphasis on identified target organs. Subchronic toxicity studies usually cannot determine the carcinogenic potential of a test substance. Guidance specific to subchronic toxicity studies with non-rodents is presented in chapter IV.C.4.b. Sponsors/submitters of petitions/notifications are encouraged to also become familiar with the Guidance for Reporting Results of Toxicity Studies (Chapter IV.B.2.), Pathology Considerations in Toxicity Studies (Chapter IV.B.3.), and Statistical Considerations in Toxicity Studies (Chapter IV.B.4.) during the development of study design.
Scientifically justified changes to the 1993 draft "Redbook II" version of this section have been made following consultation with other authoritative guidelines and publications1-8.
I. Good Laboratory Practice
Nonclinical laboratory studies must be conducted according to U.S. FDA good laboratory practice (GLP) regulations, issued under Part 58. Title 21. Code of Federal Regulations. This document may be obtained from the Superintendent of Documents, U.S. Government Printing Office, Washington, D.C., 20402, (toll free 866-512-1800).
II. Test Animals
A. Care, Maintenance and Housing:
Recommendations about the care, maintenance, and housing of animals contained in NIH publication 85-23, "Guide for the Care and Use of Laboratory Animals"8, and the DHEW publication no. 78-23 should be followed unless they conflict with specific recommendations in these guidelines.
B. Selection of Species, Strains and Sex:
These guidelines are for studies with rodents (usually rats and mice); if other species are used, modifications of these guidelines may be necessary. Both male and female test animals, that are healthy and have not been subjected to previous experimental procedures should be used.
It is important to consider the test animals' general sensitivity and the responsiveness of particular organs and tissues of test animals to toxic chemicals when selecting rodent species, strains, and substrains for toxicity studies. The selection of the use of inbred, out-bred, or hybrid rodent strains for toxicity studies should be based upon the scientific questions to be answered. Additionally, it is important that test animals come from well-characterized and healthy colonies. Because recent information suggests survivability problems exist for some strains of rats, test animals should be selected that are likely to achieve the recommended duration of the study. FDA encourages petitioners and notifiers to consult with Agency scientists before toxicity testing is begun if they have questions about the appropriateness of a particular species, strain, or substrain.
Testing should be performed on young animals, with dosing beginning as soon as possible after weaning, following an acclimation period of at least 5 days, and for rodents no later than 6 to 8 weeks of age.
D. Number and Sex:
Equal numbers of males and females of each species and strain should be used for the study. In general, for subchronic toxicity studies, experimental and control groups should have at least 20 rodents per sex per group. Ten rodents/sex/group may be acceptable for subchronic rodent studies when the study is considered to be range-finding in nature or when longer term studies are anticipated. These recommendations will help ensure that the number of animals that survive until the end of the study will be sufficient to permit a meaningful evaluation of toxicological effects.
If interim necropsies are planned, the number of animals per sex per group should be increased by the number scheduled to be sacrificed before completion of the study; for rodents, at least 10 animals per sex per group should be available for interim necropsy.
E. Infected Animals:
Generally, it is not possible to treat animals for infection during the course of a study without risking interaction between the compound used for treatment and the test substance. This interaction may confound or complicate the interpretation of study results.
F. Animal Identification:
Test animals should be characterized by reference to their species, strain (and substrain), sex, age, and weight. Each animal must be assigned a unique identification number (e.g., ear tag, implanted identification chip, tatoo).
Animals should be housed one per cage (single-caged). This recommendation reflects three principal concerns:
- The amount of feed consumed by each animal in the study cannot be determined when more than one animal is housed in each cage. This information is necessary in the determination of feed efficiency (relationship of feed consumed to body weight gained).
- Minimizing the possibility of confounding analyses and determining whether decreases in body weight gain are due to decreased palatability or substance mediated toxicity.
- Organs and tissues from moribund and dead animals which are single-caged would not be lost due to cannibalism.
In general, feed and water should be provided ad libitum to animals in toxicity studies, and the diets for these studies should meet the nutritional requirements of the species4-7 for normal growth and reproduction. Unless special circumstances apply which justify otherwise, care should be taken to ensure that the diets of the compound treated groups of animals are isocaloric (equivalent in caloric density) with and contain the same levels of nutrients (e.g., fiber, micronutrients) as the diets of the control group. Unrecognized or inadequately controlled dietary variables may result in nutritional imbalances or caloric deprivation that could confound interpretation of the toxicity study results (e.g., lifespan, background rates of tumor incidences) and alter the outcome and reproducibility of the studies.
The following issues are important to consider when establishing diets for animals in toxicity studies:
When the test substance has no caloric value and constitutes a substantial amount of the diet (e.g., more than 5%), both caloric and nutrient densities of the high dose diet would be diluted in comparison to the diets of the other groups. As a consequence, some high dose animals may receive higher test article doses than expected because animals fed such diluted diets ad libitum may eat more than animals in other dosed groups to compensate for the differences in energy and nutrient content of the high dose diets. Such circumstances make it especially important that feed consumption of these animals be as closely and accurately monitored as possible in order to determine whether changes observed could be due to overt toxicity of the test substance or to a dietary imbalance. To further aid in this assessment, two control groups can be used; one group would be fed the undiluted control diet and a second group would be fed the control diet supplemented with an inert filler (e.g., methylcellulose) at a percentage equal to the highest percentage of the test substance in the diet.
When the vehicle for the test substance is expected to have caloric and/or nutritional values, which are greater than that of the control ration, an adjustment in the caloric and/or nutritional components may be necessary.
When administration of the test substance is expected to have an effect on feed intake because of its unpleasant taste or texture, paired feeding can be used to eliminate the differences in consumption between control and compound treated groups. When a paired feeding study design is to be employed, pairs of litter-mate weanling rats of the same sex and approximate size are selected and fed the control or the experimental diet. Animals should be single-caged so that feed consumption can be determined daily, and the control animal is then fed an amount of food equal to that which the paired experimental animal ate on the preceding day. If the test substance is non-nutritive and composes a significant proportion of the diet, the pair-fed control animals should be fed an amount of its feed such that it consumes a nutritionally equivalent amount of diet as the paired experimental animal. Additionally, the study should include a second group of control animals fed ad libitum to ensure that the impact on any observed experimental result is due to differences in energy or nutrient intake.
When the test substance interferes with the absorption of nutrients, leading to nutritional deficiencies or changes in nutrient ratios, this can confound assessment of the toxicological endpoints under consideration. For example, fat soluble vitamins may preferentially partition with a mineral oil or fat substitute which is largely unabsorbed, such that a potential deficiency in these vitamins may result. This potential may be eliminated by additional nutrient fortification of the feed for those groups receiving the test substance. Appropriate levels of nutrient fortification should be determined experimentally.
It may be preferable to use a semi-purified diet prepared with known amounts of well-characterized ingredients for short-term and subchronic toxicity studies because of batch to batch variations in diet composition (e.g., fiber, mineral, vitamins, isoflavones) in some of the commonly used laboratory animal chows. The use of these semi-purified diets, however, may not be advisable in long-term and reproductive studies due to inadequate historical data related to their influences on animal survival and toxicological endpoints. For example, loss of necessary but unidentified micronutrients in the semipurified diet may interfere with normal reproduction.
Related issues are discussed in the section on Diets for Toxicity Studies in Chapter IV.B.5. in the 1993 draft "Redbook II"
I. Assignment of Control and Compound Treated Animals:
Animals should be assigned to control and compound treated groups in a stratified random manner; this will help minimize bias and assure comparability of pertinent variables across compound treated and control groups (for example: mean body weights and body weight ranges). If other characteristics are used as the basis for randomization then that characterization should be described and justified.
Animals in all groups should be placed on study on the same day; if this is not possible because of the large number of animals in a study, animals may be placed on study over several days. If the latter recommendation is followed, a preselected portion of the control and experimental animals should be placed on the study each day in order to maintain concurrence.
Excessive mortality due to poor animal management is unacceptable and may be cause to repeat the study. For example, under normal circumstances, mortality in the control group should not exceed 10%.
Adequate animal husbandry practices should result in considerably less than 10% of animals and tissues or organs lost to a study because of autolysis. Autolysis in excess of this standard may be cause to repeat the study.
Necropsy should be performed soon after an animal is sacrificed or found dead, so that loss of tissues due to autolysis is minimized. When necropsy cannot be performed immediately, the animal should be refrigerated at a temperature that is low enough to prevent autolysis but not so low as to cause cell damage. If histopathological examination is to be conducted, tissue specimens should be taken from the animals and placed in appropriate fixatives when the necropsy is performed.
III. Test Substance
The test substance used in toxicity studies should be the same substance that the petitioner/notifier intends to market. A single lot of test substance should be used throughout the study, when possible. Alternatively, lots that are similar in purity and composition should be used.
The identity of the test substance or mixture of substances to be tested should be known. We urge petitioners/notifiers to consult with the Agency in determination of test compound and to provide a Chemical Abstract Service (CAS) Registry Number or Numbers.
The composition of the test substance should be known including the name and quantities of all major components, known contaminants and impurities, and the percentage of unidentifiable materials.
C. Conditions of Storage:
The test sample should be stored under conditions that maintain its stability, quality, and purity until the studies are complete.
D. Expiration Date:
The expiration date of the test material should be known and easily available. Test materials should not be used past their expiration date.
IV. Experimental Design
A. Duration of Testing:
Animals should be exposed to the test substance 7 days per week for a minimum of 90 consecutive days (3 months). Any other regime must be justified.
B. Route of Administration:
The route of administration of the test substance should approximate that of normal human exposure, if possible. For food ingredients (e.g., food and color additives) the oral route of administration is preferred. A justification should be provided when other routes are used. The same method of administration should be used for all test animals throughout the study.
The test substance should be administered in one of the following ways:
- In the diet, if human exposure to the test substance is likely to be through consumption of solid foods or a combination of solid and liquid foods. If the test substance is added to the diet, animals should not be able to selectively consume either basal diet or test substance in the diet on the basis of color, smell, or particle size. If the compound is mixed with ground feed and pelleted, nothing in the pelleting process should affect the test substance (for example, heat-labile substances may be destroyed during pellet production by a steam process). When the test substance is administered in the diet, dietary levels should be expressed as mg of the test substance per kg of feed.
- Dissolved in the drinking water, if the test substance is likely to be ingested in liquid form (for example, in soft drinks or beer), or if administration in the diet is inappropriate for other reasons. The amount of test substance administered in drinking water should be expressed as mg of test substance per ml of water.
- By encapsulation or oral intubation (gavage), if the two previous methods are unsatisfactory or if human exposure is expected to be through daily ingestion of single, large doses instead of continual ingestion of small doses. If the test substance is administered by gavage, it should be given at approximately the same time each day. The maximum volume of solution that can be given by gavage in one dose depends on the test animal's size; for rodents, the volume ordinarily should not exceed 1 ml/100 g body weight. If the gavage vehicle is oil (see Chapter IV.B.5.b. in the 1993 draft "Redbook II"), then the volume should be no more than 0.4 ml/100 g of body weight, and the use of a low-fat diet should be considered. If the test substance must be given in divided doses, all doses should be administered within a 6 hour period. Doses of test substance administered by gavage should be expressed as mg of test substance per ml of gavage vehicle. Finally, the petitioner/notifier should provide information that can allow the reviewer to conclude that administration of the test compound by encapsulation or gavage is equivalent in all toxicologically important respects to administration in the diet or drinking water. Alternatively, metabolic information on both modes of administration should be provided so that appropriate interpretation of data can be accomplished.
C. Dose groups:
At least three dose levels of the test substance should be used per sex (one dose level per group); ideally, 4 or 5 dose levels of the test substance should be used. A concurrent control group should be included. Information from acute (Chapter IV.C.2. in the 1993 draft "Redbook II") and short-term (Chapter IV.C.3.) toxicity studies can help determine appropriate doses for subchronic studies.
1. Selection of Treatment Doses:
Dose selection for toxicity studies should be based on information related to the toxicity of the test substance.
A minimum of three dose levels of the test substance and a concurrent control group should be used in toxicity studies. When designing and conducting toxicity studies the following should be considered: 1) the high dose should be sufficiently high to induce toxic responses in test animals; 2) the low dose should not induce toxic responses in test animals; and 3) the intermediate dose should be sufficiently high to elicit minimal toxic effects in test animals (such as alterations in enzyme levels or slight decreases in body weight gains). No dose should cause an incidence of fatalities that prevents meaningful evaluation of the data. Administration of the test substance to all dose groups should be done concurrently.
A concurrent control group of test animals is required. The control group in dietary studies should be fed the basal diet. Exceptions to this and other related information, including a discussion regarding pair-feeding, was provided above in section "II Test Animals, H. Diet".
A carrier or vehicle for the test substance should be given to control animals at a volume equal to the maximum volume of carrier or vehicle given to any dosed group of animals. Sufficient toxicology information should be available on the carrier or vehicle to ensure that its use will not compromise the results of the study. If there is insufficient information about the toxic properties of the vehicle used to administer the test substance, an additional control group that is not exposed to the carrier or vehicle should be included. In all other respects, animals in the control group should be treated the same as animals in dosed groups. (See additional information in section "II Test Animals, H. Diet" above.)
D. Computerized systems
Computerized systems that are used in the generation, measurement, or assessment of data should be developed, validated, operated, and maintained in ways that are compliant with Good Laboratory Practice principles. 9
V. Observations and Clinical Tests
A. Observations of Test Animals:
Routine cage-side observations should be made on all animals at least once or twice a day throughout the study for general signs of pharmacologic and toxicologic effects, morbidity and mortality. The usual interval between observations should be at least 6 hours. Individual records should be maintained for each animal and the time of onset and the characteristics and progression of any effects should be recorded, preferably using a scoring system.
An expanded set of clinical evaluations, performed inside and outside of the cage, should be carried out in short-term and subchronic toxicity studies in rodents and non-rodents, in one-year non-rodent toxicity studies, and reproductive toxicity studies in rodents to enable detection not only of general pharmacologic and toxicologic effects but also of neurologic disorders, behavioral changes, autonomic dysfunctions, and other signs of nervous system toxicity. Specific information about the systematic clinical tests/observations is contained in Chapter IV.C.10. This expanded set of clinical examinations (Chapter IV.C.10), conducted inside and outside the cage, should be age appropriate and performed on all animals at least once prior to initiation of treatment, and periodically during treatment. Signs noted should include, but not be limited to, changes in skin, fur, eyes, mucous membranes, occurrence of secretions and excretions or other evidence of autonomic activity (e.g., lacrimation, piloerection, pupil size, unusual respiratory pattern). Additionally, changes in gait, posture and response to handling, as well as the presence of clonic or tonic seizures, stereotypes (e.g., excessive grooming, repetitive circling) or bizarre behavior (e.g., self-mutilating, walking backwards) should be recorded. Tumor development, particularly in long-term studies, should be followed: the time of onset, location, dimensions, appearance and progression of each grossly visible or palpable tumor should be recorded. During the course of a study, toxic and pharmacologic signs may suggest the need for additional clinical tests or expanded post-mortem examinations.
B. Body Weight and Feed Intake Data:
Test animals should be weighed at least once a week. Feed consumption (or water consumption if the test substance is administered in the drinking water) should be measured every week during the subchronic toxicity study. Petitioners should also attempt to quantify spillage of feed by test animals, and to determine if spillage is greater with test diets than with control diets. Appropriate discussions of feed spillage should be included in the study report.
C. Clinical Testing:
Ophthalmological examination, hematology profiles, clinical chemistry tests, and urinalyses should be performed as described in the following sections:
- Opthalmological Examination: This examination should be performed by a qualified individual on all animals before the study begins and on control and high-dose animals at the end of the study. If the results of examinations at termination indicate that changes in the eyes may be associated with administration of the test substance, ophthalmological examinations should be performed on all animals in the study.
Hematology: Blood samples should be obtained from a minimum of 10 rodents of each sex per group at least three times during the study. Sampling of test animals should be conducted during the first two weeks on study (receiving treatment), monthly or midway through treatment (day 45), and at termination. The determination of the first sampling time point should be based on the expected time of initial toxicological effects on the organ systems. Ideally, the same rodents should be sampled during the study, and at termination. The collection of blood samples should occur at approximately the same time on each sampling day. If animals are fasted prior to sampling, then blood collection should occur at the conclusion of the fast and prior to feeding. Fasting duration should be appropriate for the species and the analytical tests to be performed. Hematologic tests should be performed on individual samples and not pooled.
- The following determinations are recommended:
- hemoglobin concentration
- erythrocyte count
- total and differential leukocyte counts
- mean corpuscular hemoglobin
- mean corpuscular volume
- mean corpuscular hemoglobin concentration
- and a measure of clotting potential (such as clotting time, prothrombin time, thromboplastin time, or platelet count).
Test compounds may have an effect on the hematopoietic system and therefore appropriate measures should be employed so that evaluations of reticulocyte counts and bone marrow cytology may be performed if warranted. Reticulocyte counts should be obtained for each animal using automated reticulocyte counting capabilities, or from air-dried blood smears. Bone marrow slides should be prepared from each animal for evaluating bone marrow cytology. These slides would only need to be examined microscopically if effects on the hematopoietic system were noted.
- The following determinations are recommended:
Clinical Chemistry: Blood samples should be obtained from a minimum of 10 rodents of each sex per group at least three times during the study. Sampling of test animals should be conducted during the first two weeks on study (receiving treatment), monthly or midway through treatment (day 45), and at termination. The determination of the first sampling time point should be based on the expected time of initial toxicological effects on the organ systems. Ideally, the same rodents should be sampled during the study, and at termination. The collection of blood samples should occur at approximately the same time on each sampling day. If animals are fasted prior to sampling, then blood collection should occur at the conclusion of the fast and prior to feeding. Fasting duration should be appropriate for the species and the analytical tests to be performed. Clinical chemistry tests should be performed on individual samples and not pooled.
Clinical chemistry tests that are appropriate for all test substances include measurements of electrolyte balance, carbohydrate metabolism, and liver and kidney function. Specific determinations should include:
- Hepatocellular evaluation: select at least 3 of the following 5
- alanine aminotransferase (SGPT, ALT)
- aspartate aminotransferase (SGOT, AST)
- sorbitol dehydrogenase
- glutamate dehydrogenase
- total bile acids
- Hepatobiliary evaluation: select at least 3 of the following 5
- alkaline phosphatase
- bilirubin (total)
- gamma-glutamyl transpeptidase (GG transferase)
- 5' nucleotidase
- total bile acids
- Other markers of cell changes or cellular function
- globulin (calculated)
- glucose (in fasted animals)
- protein (total)
- triglycerides (fasting)
- urea nitrogen
However, when adequate volumes of blood cannot be obtained from test animals, the following determinations should generally be given priority. FDA understands that the specific nature of the test compound may warrant the consideration of alternative tests. Appropriate justification for alternative tests should be presented in study reports.
- alanine aminotransferase
- alkaline phosphatase
- gamma-glutamyl transpeptidase (GG transferase)
- glucose (in fasted animals)
- protein (total)
- urea nitrogen
Additional clinical chemistry tests may be recommended to extend the search for toxic effects attributable to a test substance. The selection of specific tests will be influenced by observations on the mechanism of action of the test substance. Clinical chemistry determinations that may be recommended to ensure adequate toxicological evaluation of the test substance include analyses of acid/base balance, hormones, lipids, methemoglobin, and proteins.
In spite of standard operating procedures and equipment calibration, it is not unusual to observe considerable variation in the results of clinical chemistry analyses from day to day10. Ideally, clinical chemistry analyses for all dose groups should be completed during one day. If this is not possible, analyses should be conducted in such a way as to minimize potential variability.
- Hepatocellular evaluation: select at least 3 of the following 5
- Urinalyses: Timed urine volume collection should be conducted during the last week of the study on at least 10 animals of each sex in each group. The volume of urine collected, specific gravity, pH, glucose, and protein should be determined as well as conducting a microscopic evaluation of urine for sediment and presence of blood/blood cells11.
Neurotoxicity Screening/Testing : Screening for neurotoxic effects should be routinely carried out in all subchronic toxicity studies with rodents (preferably rats). The neurotoxicity screen should be age appropriate and would typically include: (1) specific histopathological examination of tissue samples representative of major areas of the brain, spinal cord, and peripheral nervous system (see organs and tissues listed below under VI.C. Preparation of Tissues for Microscopic Examination) and (2) a functional battery of quantifiable observations and manipulative tests selected to detect signs of neurological, behavioral, and physiological dysfunctions. This functional battery is also referred to as an expanded set of clinical evaluations and is described more fully in section V.A. Observations of Test Animals in this chapter and in Chapter IV.C.10. Neurotoxicity Studies.
Subchronic toxicity study reports should include an assessment of the potential for the test substance to adversely affect the structural or functional integrity of the nervous system. This assessment should evaluate data from the neurotoxicity screen and other toxicity data from the study, as appropriate. Based on this assessment, the petitioner should make an explicit statement about whether or not the test substance presents a potential neurotoxic hazard and if additional neurotoxicity testing is deemed appropriate. FDA recommends that additional neurotoxicity testing not be undertaken without first consulting with the Agency.
- Immunotoxicity: Results from tests that are included in the list of primary indicators for immune toxicity (see Chapter V.C. of the 1993 draft "Redbook II") should also be evaluated as part of an immunotoxicity screen. Reports of subchronic toxicity studies should include an assessment of the potential for the test substance to adversely affect the immune system. This assessment should evaluate data from the list of primary indicators included in the immunotoxicity screen and other toxicity data from the study, as appropriate. Based on this assessment, the petitioner/notifier should make an explicit statement about whether or not the test substance presents a potential immunotoxic hazard which requires further immunotoxicity testing. Additional immunotoxicity tests are discussed in Chapter V. C. of the 1993 draft "Redbook II", but should not be undertaken without first consulting with the Agency.
VI. Necropsy and Microscopic Examination
A. Gross Necropsy
All test animals should be subjected to complete gross necropsy, including examination of external surfaces, orifices, cranial, thoracic and abdominal cavities, carcass, and all organs. The gross necropsy should be performed by, or under the direct supervision of, a qualified pathologist, preferably the pathologist who will later perform the microscopic examination (see below).
B. Organ Weight
Organs that should be weighed include the adrenals, brain, epididymides, heart, kidneys, liver, spleen, testes, thyroid/parathyroid, thymus, ovaries and uterus. Organs should be carefully dissected and trimmed to remove fat and other contiguous tissue and then be weighed immediately to minimize the effects of drying on organ weight.
C. Preparation of Tissues for Microscopic Examination
Generally, the following tissues should be fixed in 10% buffered formalin (or another generally recognized fixative) and sections prepared and stained with hematoxylin and eosin (or another appropriate stain) in preparation for microscopic examination. Lungs should be inflated with fixative prior to immersion in fixative.
- bone (femur)
- bone marrow (sternum)
- brain (at least 3 different levels)
- corpus and cervix uteri
- gall bladder (if present)
- Harderian gland (if present)
- lung (with main-stem bronchi)
- lymph nodes (1 related to route of administration and 1 from a distant location)
- mammary glands
- nasal turbinates
- ovaries and fallopian tubes
- salivary gland
- sciatic nerve
- seminal vesicle (if present)
- skeletal muscle
- spinal cord (3 locations: cervical, mid-thoracic, lumbar)
- thymus (or thymic region)
- urinary bladder
- all tissues showing abnormality
D. Microscopic Evaluation
All gross lesions should be examined microscopically. All tissues from the animals in the control and high dose groups should be examined. If treatment related effects are noted in certain tissues, then the next lower dose level tested of those specific tissues should be examined. Successive examination of the next lower dose level continues until no effects are noted. In addition, all tissues from animals which died prematurely or were sacrificed during the study should be examined microscopically to assess any potential toxic effects.
E. Histopathology of Lymphoid Organs
Histopathological evaluation of the lymphoid organs should be performed as described in the section on immunotoxicity testing (see Chapter V. C. of the 1993 draft "Redbook II") for all animals.
- EPA Health Effects Test Guidelines OPPTS 870.3100, 90-Day Oral Toxicity in Rodents, August 1998
- OECD Guideline For The Testing Of Chemicals , Repeated Dose 90-day Oral Toxicity Study in Rodents, 408, September 1998
- Kurt Weingand et al., 1996. "Harmonization of Animal Clinical Pathology Testing in Toxicity and Safety Studies", Fundamental and Applied Toxicology, 29, pp 198-201.
- Nutrient Requirements of Laboratory Animals, Fourth Revised Edition, Subcommittee on Laboratory Animal Nutrition, Committee on Animal Nutrition, Board on Agriculture, National Research Council, 1995
- Nutrient Requirement of Dogs, Revised, Committee on Animal Nutrition, National Research Council, 1985.
- Nutrient Requirements of Swine, 10th Revised Edition, Subcommittee on Swine Nutrition, Committee on Animal Nutrition, National Research Council, 1998.
- Nutrient Requirements of Rabbits, Second Revised Edition, Committee on Animal Nutrition, National Research Council, 1977.
- National Research Council Institute of Laboratory Animal Resources. 1996. Guide for the Care and Use of Laboratory Animals. National Academy Press, Washington, DC.
- The Application of the Principles of GLP to Computerised Systems, Organisation for Economic Cooperation and Development (OECD), Environmental Monograph No. 116, Paris, 1995.
- Gaylor, D.W., Suber, R.L., Wolff, G.L. and Crowell, J.A. (1987) Statistical variation of selected clinical pathological and biochemical measurements in rodents (42555) Proc. Soc. Exp. Biol. Med. , 185:361-367
- Ragan, H.A. and Weller, R.E., "Markers of Renal Function and Injury" in The Clinical Chemistry of Laboratory Animals, Second Edition 1999, Taylor&Francis, Philadelphia, PA, pp. 520-533