Complete Blood Count, Clinical Chemistry, and Serology Profile by Using a Single Tube of Whole Blood from Mice

Charles E Wiedmeyer,1,* Dawn Ruben,3 and Craig Franklin2

Clinical pathology is a valuable means for assessing specific organ pathology and a screening tool for general animal health. Routine clinical pathology evaluation in mice usually includes whole blood for a complete blood count (CBC) and a clinical biochemistry analysis. Acquisition and analysis of these samples can be problematic due to the small volumes of blood that can be obtained from a mouse. Typically, a complete blood count requires blood from a tube containing an anticoagulant, whereas a clinical biochemistry profile needs blood from a serum clot tube. Because of the small volume that can be obtained, splitting the blood from a single mouse into 2 different tubes may result in inadequate samples to perform the desired tests or introduce inaccuracies. We explored the feasibility of using a single lithium heparin tube for generation of a CBC, biochemistry profile, and serology profile. We also evaluated the consistency of CBC data, including the quality of a peripheral blood smear taken from a lithium heparin or EDTA tube after various storage times. We found that CBC, biochemistry, and serology profiles could be obtained more readily when blood samples were placed in a single lithium heparin tube than in 2 separate tubes. In addition, the quality of blood smears and CBC results from the lithium heparin tube were comparable (with few exceptions) to those from an EDTA tube after prolonged storage.

Abbreviations: CBC, complete blood count; EDTA, ethylenediaminetetraacetic acid; HCT, hematocrit; MCH, mean corpuscular hemoglobin; MCHC, mean corpuscular hemoglobin concentration; MCV, mean corpuscular volume; RBC, red blood cells; WBC, white blood cells

Clinical pathology evaluation of mice used for toxicity and Although clinical pathology testing may be very useful, the safety studies or biomedical research is relatively common and acquisition and analysis of samples from mice is problematic required by various regulatory agencies.2,4,5 Clinical pathol- due to the small sample volumes that can be obtained from a ogy data can identify disease in specific target organ systems, single animal.6,9,14,15 Pooling of samples is not recommended, provide a general health profile of the individual animal, and and interim (in life) testing is usually not possible for complete establish dose-response relationships.3 Guidelines for clinical clinical pathology evaluation of a single animal.2,15 Most clinical pathology testing on animals used for toxicity and safety studies pathology samples, especially whole blood and bone marrow have been established.4,15 The parameters recommended for a smears, are taken at termination. Typically, blood taken from core hematology evaluation include: a total leukocyte (WBC) the mouse at termination is allocated into 2 different tubes.6 One count, absolute differential leukocyte count, erythrocyte (RBC) tube, containing an anticoagulant such as ethylenediaminetet- count, hematocrit (HCT) or packed cell volume, hemoglobin raacetic acid (EDTA), is used for a complete blood count, and concentration, mean corpuscular volume (MCV), mean cor- the other, a serum clot tube, for serum chemistry analysis. For puscular hemoglobin (MCH), mean corpuscular hemoglobin coagulation studies, blood must be added to another specific concentration (MCHC), erythrocyte morphology, and platelet tube (that is, citrate).6,10 The use of 2 (or more) different blood count. A blood smear for a reticulocyte count and a bone marrow tubes may result in an inadequate amount of blood in each tube; smear for cytologic examination also are recommended.2,3,15 The an inadequate amount of blood in the tube may yield inaccurate recommended clinical chemistry profile includes: glucose, urea or unobtainable results.1,14 nitrogen, creatinine, total protein, albumin, calculated globulin, Moreover, mice requiring clinical pathology often are shipped calcium, sodium, potassium, cholesterol, and hepatocellular to diagnostic laboratories for necropsy and terminal blood and hepatobiliary tests.2,3,15 For hepatocellular evaluation, a collections. During shipping, mice typically are provided a minimum of 2 appropriate tests is recommended; these in- source of hydration other than water (that is, hydration gel). clude alanine aminotransferase, aspartate aminotransferase, This practice may lead to mild, subclinical dehydration that sorbitol dehydrogenase, glutamate dehydrogenase, and total further precludes obtaining sufficient blood volume for clinical bile acids.2,3,15 The hepatobiliary tests that may be evaluated pathology. Alternatively, rather than shipping mice for necropsy include alkaline phosphatase, gamma glutamyltransferease, 5' and blood removal, blood samples taken at the home facility nucleotidase, total bilirubin, and total bile acids.2,3,15 are shipped on ice to the laboratory. Therefore, transit time and temperature may alter the quality of clinical pathology results Received: 30 Jun 2006. Revision requested: 6 Sep 2006. Accepted: 19 Sep 2006. obtainable from these samples. 1Veterinary Medical Diagnostic Laboratory, College of Veterinary Medicine, and 2Research In this study we compared the use of a single heparinized Animal Diagnostic Laboratory, Department of Veterinary Pathobiology, University of tube for generation of an automated complete blood count, a Missouri-Columbia, Columbia, MO; 3Department of Molecular Comparative Pathobiol- ogy, Johns Hopkins School of Medicine, Baltimore, MD. 14-parameter chemistry profile, and a 19-parameter serology *Corresponding author. Email: [email protected] profile to using 2 separate tubes (that is, EDTA and serum clot

Vol 46, No 2 Journal of the American Association for Laboratory Animal Science March 2007

Figure 1. Technique used for mouse cardiocentesis to obtain all blood samples. tubes). We also assessed whether the use of gel packs as a water source affects the performing of these tests. Finally, the quality of CBC data, including that of smears, from blood placed in 2 different anticoagulants and stored for various lengths of Figure 2. Tubes used for this study. (A) Red-top serum tube. (B) EDTA tube. (C) Lithium heparin separator tube. time was examined. Our results indicate that providing mice with water rather than a gel pack for hydration and using a single tube for blood collection increase the amount of clinical water bottle, and 10 male and 10 female mice received a gel pack pathology data that can be obtained from a single mouse. CBC as a water source. For this study, mice were housed in same-sex data were comparable (with few exceptions) between blood groups of 5 under the same conditions as stated previously. stored in lithium heparin and EDTA at 4 pC after various inter- All mice were euthanized by CO2 overdose and weighed, and vals. However, the quality of blood smears deteriorated with whole blood samples taken via cardiocentesis with a 22-gauge prolonged storage; therefore smears should be created shortly needle immediately after euthanasia (Figure 1). Whole blood after blood collection. taken from 10 (5 male and 5 female) mice in each group was placed in 0.5-ml lithium heparinized gel separator tubes (Becton -ATERIALS¬AND¬-ETHODS Dickinson, Franklin Lakes, NJ; Figure 2) by removing the needle !NIMALS The population for the tube study comprised 20 from the syringe, placing the blood into the tube, and mixing male and 20 female DBA/2 mice (Mus musculus) from an in- thoroughly. The lithium heparin tubes used do not contain a house source. Mice were housed in rooms that were monitored visible amount of anticoagulant and are supplied with a gel by use of a quarterly sentinel mouse program. Sentinels were plug for facilitation of plasma separation from cells. Therefore, consistently free of mouse hepatitis virus, mouse minute virus, dilution of the blood sample with anticoagulant is negligible. mouse parvovirus, Sendai virus, Mycoplasma pulmonis, Theiler Whole blood from the other 10 mice in each group was treated in murine encephalomyelitis virus, epizootic diarrhea of infant the same manner but was used first to fill an EDTA tube (Becton mice, pneumonia virus of mice, reovirus type 3, lymphocytic Dickinson, Franklin Lakes, NJ) to the manufacturer-required choriomeningitis virus, ectromelia virus, and ecto- and endopar- amount (500 Nl),13 and the remaining amount of blood was asites. The study was conducted in accordance with guidelines placed in a red-top serum clot tube (Becton Dickinson; Figure set forth by the Guide for the Care and Use of Laboratory Animals8 2). Each sample was analyzed within 2 h of acquisition. and approved by the University of Missouri-Columbia Animal Shortly after the sample was taken, 1 femur from each ani- Care and Use Committee. Mice were housed in polycarbonate mal was removed and the marrow flushed from the bone by static microisolation cages (Thoren, Hazelton, PA) with auto- used of formalin-buffered saline and a syringe. The flushed claved pelleted bedding (Paperchip, Canbrands International, bone marrow was smeared on several glass slides by use of a Moncton, New Brunswick, Canada) and were allowed ad libi- squash technique. tum access to Irradiated Picolab 20, 5053 Rodent Chow (Purina Whole blood from the 18 mice for the blood-storage ex- Mills, Richland, IN) and autoclaved tap water. Environmental periment was collected in the same manner as previously conditions consisted of 10 to 15 air changes hourly, room tem- mentioned. The whole blood from 9 of the mice was pooled in perature of 21 q 2 pC, relative humidity of 30% to 70%, and a a 10.0-ml lithium heparinized tube, and blood from the other 9 12:12-h light:dark cycle. The mice were part of an unrelated mice was placed in a 10.0-ml EDTA tube. The 9 mice used for study and were slated for termination. The mice ranged in age each pooled sample were not discriminated by sex. The tubes from 4 to 5 mo and weight from 19.0 to 37.0 g. were mixed thoroughly, and 500-Nl aliquots were placed in An additional 18 (9 male and 9 female) DBA/2 mice, housed respective anticoagulant tubes. under the same conditions described earlier, were used to ex- 3AMPLE¬ANALYSIS¬For the tube experiment, each whole blood amine blood smear quality and CBC data after various times sample placed in the heparinized tube was analyzed in the of blood storage. same fashion and in a specific order. The order of tests was: 1) 3AMPLE¬ACQUISITION¬The mice used for the tube experiment a peripheral blood smear was made; 2) an automated CBC was were allocated into 4 evenly distributed groups: 10 male and 10 performed; 3) tubes were centrifuged at 10,000 t g for 5 min in female mice were kept overnight (approximately 16 h) with a a benchtop centrifuge (IEC Micro-MB, Thermodyne, Needham

Clinical pathology of mice by using a single tube of blood

Table 1. Results using lithium heparin single- and 2-tube methods in mice supplied overnight with water or a gel pack for hydration Mean (range) Mean (range) No. with full chemistry No. with full serology Method weight (g) blood volume (Nl) profile profile Male mice, water 1 tube 32.6 (30–37) 800 (800) 5 5 Male mice, water 2 tube 33.1 (22.5–37) 670 (350–800) 0 3 Male mice, gel pack 1 tube 25.0 (21–33) 690 (550–800) 3 5 Male mice, gel pack 2 tube 33.0 (31–35) 680 (500–800) 0 3 Female mice, water 1 tube 21.0 (19–22) 590 (500–800) 4 5 Female mice, water 2 tube 22.2 (21–23) 630 (400–800) 0 2 Female mice, gel pack 1 tube 23.6 (22–26) 670 (550–800) 4 5 Female mice, gel pack 2 tube 22.4 (21–25) 600 (500–700) 0 0 Each group contained 5 mice, and full complete blood counts were obtained on all 5 mice in each group.

Heights, MA); 4) plasma was withdrawn from the tube and ana- adequate cytologic specimens for bone marrow examination. lyzed for a clinical chemistry profile; and 5) remaining plasma was assayed for a serology profile. For samples placed in the 2ESULTS EDTA tubes, the same procedure used to generate a CBC was 4UBE¬STUDY Results of this experiment are listed in Table 1. The performed but the sample was not used for chemistry analysis. amount of whole blood obtained from all the animals ranged The serum clot tube was centrifuged in the same manner, and from 350 to 800 Nl. The average amount of whole blood taken all the serum was removed from the tube. from the overnight water bottle group was 672 Nl (range, 350 For the storage study, two 500-Nl aliquots of blood were used to 800 Nl) compared with 660 Nl (range, 550 to 800 Nl) from the at each time point and the results averaged. Blood from each group supplied a gel pack overnight. A full CBC could be ob- heparin and EDTA tube was used to create peripheral blood tained from all of the mice. By using the plasma obtained with smears and underwent automated CBC analysis at the time of the single lithium heparin tube method, a full 14-parameter collection (time 0) and at 0.5, 24, 48, and 72 h after collection. All clinical chemistry profile could be obtained from 9 of the 10 (5 aliquots of blood were stored at 4 pC between analyses. of 5 male, 4 of 5 female) mice supplied water overnight and 7 Hematology. Prior to CBC analysis with an automated hema- of the 10 (3 of 5 male, 4 of 5 female) mice provided gel packs tology instrument (Abbott Cell-Dyn 3500 Hematology Analyzer, for hydration. Of the 4 plasma samples obtained using the 1- Abbott Labs, Abbott Park, IL), a drop of blood was withdrawn tube method in which a full profile was not achieved, at least 11 from the tube to create a peripheral blood smear. The peripheral chemistry parameters were obtained. From the serum samples blood smear was stained with modified Wright Geimsa by use taken from the mice whose blood was split into the 2 tubes, a of an automated slide stainer (Hematek, Miles, Elkhart, IN). full clinical chemistry profile could not be obtained regardless The peripheral blood smear was used to determine RBC and of the water source. An average of 8 parameters (range, 0 to 12) WBC morphology, estimate platelet number and distribution, in the chemistry profile was obtained from these samples. A full and confirm automated leukocyte counts. Automated CBC 19-parameter serology profile was obtained from all 10 mice parameters included: WBC count, absolute differential WBC by using the 1-tube method and supplying water, all 10 mice count, RBC count, HCT, hemoglobin concentration, MCV, MCH, by using the 1-tube method and supplying gel packs, 5 of 10 (3 MCHC, and platelet count. of 5 male, 2 of 5 female) mice by using the 2-tube method and Chemistry proﬁle. After the CBC and centrifugation were supplying water, and 3 of 10 (3 of 5 male, 0 of 5 female) mice by completed, the plasma or serum was taken from the tube and using the 2-tube method and supplying gel packs. Using data placed in a microfuge tube for clinical chemistry analysis by an from all mice, we established that a minimal weight of 19.0 g automated analyzer (Olympus AU400, Olympus America, Ir- and minimal volume of 500 Nl blood were necessary to obtain ving, TX). The following chemistry parameters were determined a CBC, full clinical chemistry profile, and serology. Adequate on each serum or plasma sample: glucose, blood urea nitrogen, bone marrow samples were obtained from all mice. creatinine, sodium, potassium, chloride, albumin, total protein, 3TORAGE¬ STUDY CBC results from blood stored in lithium calculated globulin, calcium, cholesterol, total bilirubin, gamma heparin and EDTA showed few noteworthy changes over time glutamyltransferase, and alanine aminotransferase. (Table 2). The greatest change in the lithium heparin tube was a Serology proﬁle. Plasma or serum remaining after the chem- decrease in platelet numbers, as determined by the automated istry analysis underwent a 19-parameter serologic profile. CBC instrument, after 24 h of storage and a gradual increase at Serology was performed by use of multiplex fluorescent im- 48 and 72 h. (Figure 3) Blood stored in EDTA showed a similar 12 munoassay technology (Luminex 100, Luminex, Austin, TX). pattern of decline in platelet numbers after 24 h of storage, Antibody levels to the following organisms were determined: with increased numbers at 48 and 72 h (but still less than the cilia-associated respiratory bacillus, Encephalitozoon cuniculi, ec- counts at time 0 and 0.5 h; Figure 3). The WBC decreased and tromelia, epizootic diarrhea of infant mouse virus, lymphocytic HCT increased over time in the lithium heparin blood (Table 2); choriomeningitis virus, Mycoplasma pulmonis, mouse adenovirus these changes were not seen in the EDTA-treated blood. Visual 1, mouse adenovirus 2, mouse cytomegalovirus, mouse hepatitis examination of blood smears from each of the peripheral blood virus, mouse parvovirus, minute virus of mice, mouse recom- smears made after storage did not reveal marked morphologic binant parvoviral protein, polyoma virus, pneumonia virus of changes in the RBC. Platelets were distributed evenly on the mice, reovirus type 3, Sendai virus, Theiler murine encephalo- slide at times 0 and 0.5 h in both the lithium and EDTA samples, myelitis virus, and Clostridium piliforme. but large platelet clumps appeared at the feathered edge of the

Bone marrow examination. Prepared bone marrow samples smears of both at later time points (Figure 4). WBC morphol- were air dried and stained with modified Wright Geimsa by ogy did not vary greatly throughout the storage period in both use of an automated stainer. The slides were confirmed to be

Vol 46, No 2 Journal of the American Association for Laboratory Animal Science March 2007

Table 2. Results of complete blood counts from pooled blood samples stored in lithium heparin or EDTA tubes for 0, 0.5, 24, 48, and 72 h Neutro- Lympho- Eosino- Hemo- Hemato- WBC Monocytes Basophils RBC MCV MCH MCHC Platelets phils cytes phils globin crit (t103/Nl) (t103/Nl) (t103/Nl) (t106/Nl) (fl) (pg) (g/dl) (t103/Nl) (t103/Nl) (t103/Nl) (t103/Nl) (g/dl) (%)

EDTA, 0 h 8.99 0.99 7.52 0.19 0.03 0.25 9.57 14.5 45.1 47.1 15.2 32.2 1066 EDTA, 0.5 h 8.58 0.93 7.11 0.24 0.03 0.27 9.49 14.5 44.9 47.4 15.3 32.3 1011 EDTA, 24 h 9.81 0.99 8.21 0.21 0.02 0.36 9.49 14.6 45.2 47.6 15.4 32.3 854 EDTA, 48 h 9.99 1.02 8.42 0.18 0.02 0.38 9.38 14.6 44.5 47.5 15.6 32.8 940 EDTA, 72 h 9.52 0.96 7.87 0.28 0.02 0.39 9.52 14.8 45.4 47.8 15.5 32.5 934 Heparin, 0 h 7.41 1.33 5.53 0.19 0.01 0.35 9.94 11.3 46.8 47.2 11.4 24.2 1000 Heparin, 0.5 h 7.14 1.40 5.42 0.47 0.01 0.20 9.72 15 45.9 47.2 15.5 32.7 942 Heparin, 24 h 4.53 0.79 3.44 0.17 0.01 0.24 10.05 15.3 47.8 47.5 15.2 32.0 96 Heparin, 48 h 5.31 1.01 3.81 0.07 0.01 0.24 10.24 15.7 48.6 47.6 15.4 32.3 187 Heparin, 72 h 4.93 1.54 3.10 0.08 0.02 0.21 11.1 16.9 52.5 47.3 15.3 32.4 441 samples, but increased numbers of ruptured and disrupted WBCs were present in the 72-h samples of both.

$ISCUSSION¬ The mouse is the most widely used mammal for biomedical research and is often used in safety and toxicity studies.9 Gen- eral clinical pathology data are used to monitor organ function, determine dose-response relationships, and as a measure of overall health. The use of clinical pathology in mice has been limited due to their small size and the volume of blood that can be extracted from a single animal.2,9,14,15 A 35-g adult mouse may have a total blood volume of approximately 2.4 ml, but this volume varies with the size and sex of the mouse.6,9 Numerous techniques for obtaining a blood sample from a mouse have been 6 described. Most of these techniques are designed to obtain an Figure 3. Numbers of platelets in blood stored in EDTA or lithium interim (in life) sample and, although useful for repeated blood heparin for 0, 0.5, 24, 48, and 72 h. draws, they typically yield an insufficient amount of blood to perform both a CBC and chemistry profile, as well as other tests. decrease the number of animals needed for many studies. Therefore, if a CBC, complete chemistry profile, and other tests As stated previously, obtaining an adequate volume of blood requiring a substantial amount of blood are necessary, interim to perform both hematology and chemistry testing can be prob- blood samples from mice are not recommended and likely are lematic due to the mouse’s small size. One source of error that not possible.15 In addition, pooling of samples is not a recom- may be introduced and prevent these 2 profiles from reliably mended practice.15 being performed is the need for the blood to be placed in 2 dif- Techniques that have shown to yield greater amounts of blood ferent tubes. For a complete blood count and a chemistry profile have been described.9 These techniques require anesthesia and to be determined from a single mouse, the whole blood drawn catheterization of a major vessel or exposure of the heart and is split into 2 tubes: an EDTA tube for a CBC and a clot tube for are terminal.9 For the present study, we used cardiocentesis to a clinical chemistry profile. A CBC must be performed from a eliminate the need for vessel catheterization or exposure of the tube that contains an anticoagulant, and EDTA is the preferred heart. We found that this method, when done properly, can anticoagulant for routine hematologic testing.10 Because of yield as much as 800 Nl blood, which usually was sufficient EDTA’s mechanisms of action and active form, EDTA-treated to perform a CBC, chemistry profile, and a serology profile by blood should not be used for clinical chemistry analysis because using a single tube. of error in measurement of several electrolyte parameters (that Recommendations by various regulatory agencies for clini- is, potassium, sodium, and calcium).10 cal pathology testing on mice usually require hematology and Heparin is an acceptable anticoagulant for routine hemato- clinical chemistry testing, depending on the study type.2-4 The logic testing and has the advantage that most clinical chemistry parameters recommended for inclusion in hematology and clini- assays can be performed on heparin-preserved blood.10 Lithium cal chemistry testing have been outlined previously.2-4 Although heparin tubes are recommended, because sodium heparin tubes the recommendation of obtaining CBC and clinical chemistry will result in abnormal electrolyte values when the plasma from profiles from the same animal is not stated explicitly, having these tubes is analyzed. In addition, cellular morphology is pre- this information from a single animal may help in identifying served longer in heparinized samples.6 Citrate and oxalates are a greater breadth of pathologies and thus provide more useful other anticoagulants that can be used for CBCs but are usually data for the study as well as contribute to characterizing phe- reserved for specialized testing. notypes. Moreover, the ability to obtain both data sets from a In addition, the practice of splitting a sample into 2 tubes single animal compared with multiple animals would have con- may introduce error. Because the sample must be split into 2 siderable ethical and economic impact, because doing so would fractions, each tube may receive an inadequate amount of blood.

Clinical pathology of mice by using a single tube of blood

both CBC and clinical chemistry data circumvents the potential problems that may be associated with using 2 tubes. The results of this study reveal that an appropriate blood sampling technique, blood tube, and sequence of procedures must be used with the 1-tube procedure. As described earlier, cardiocentesis at termination usually provides sufficient blood for a CBC, chemistry profile, and serology profile. A lithium heparinized blood tube is needed to avoid clotting of the sample and circumvent erroneous chemistry results that occur when using EDTA. To maximize the amount of plasma from the sample and to prevent contamination with blood cells when the plasma is extracted from the tube, a separator tube that contains a gel plug provides the best results to accomplish clean separation of cells from plasma. The sequence of the procedures performed from the single tube is vital for accurate results. After the blood is introduced into the tube and mixed thoroughly, the first procedure consists of taking a drop of blood to make a blood smear. Second, the blood is analyzed for a CBC by using an automated hematology instrument. Third, the tube is centrifuged to separate the cells from the plasma. It is important to extract as much plasma as possible from the tube in order to obtain an adequate volume of plasma for the chemistry and serology profiles. Again, the use of the separator tube with a gel plug is ideal for obtaining maximal plasma from the sample. Fourth, the plasma is analyzed for the chemistry profile by using the automated instrument, and the remaining plasma is used for the serology profile. We found that sufficient plasma remained in all 20 samples after this 1- tube procedure to obtain a full serology profile. By using the 2-tube method, only 8 of the 20 samples yielded enough serum to perform a full serology profile after the chemistry analysis. The remaining 12 samples either contained insufficient sample volume to perform the serology profile, or the quality of the sample after chemistry analysis was poor (that is, slightly clot- ted). Therefore, we recommend the use of the 1-tube to obtain sufficient plasma to perform a full chemistry and serology profile. In addition, as a general rule, we recommend performing the blood smear and CBC and extracting the plasma within 2 h of acquisition to obtain optimal results. Possible complication of delayed analysis are platelet clumping and decreased plasma glucose; plasma glucose will decrease approximately 10% for every hour the plasma is not separated from the RBCs.10 Once the plasma is separated from the RBC, the chemistry profile and serology may be done later, but other complications may arise after refrigerating or freezing the plasma samples. In addition to hematology and clinical chemistries, we also obtained a serology profile by using the 1-tube method, likely due to the yield of serum and the inherent low volume require- ment of multiplex fluorescence immunoassay methodology.12 Multiplex fluorescence immunoanalysis is a bead-based test that requires less than 1.0 Nl serum for testing multiple ( 20) agents. Serum samples using the 2-tube method often yielded lower Figure 4. Photomicrographs of peripheral blood smears from lithium volumes, and a full serology profile could not be performed heparin tube. Smears were made at (A) time 0 and (B and C) after storage in 12 of 20 samples despite the low sample volume required of blood for 24 h. The black arrows indicate that platelets are distributed evenly in the time-0 smear but are clumped (B) and patchily distributed for this test. (C) in smears from stored blood. Our primary goal was to determine whether the 1-tube method is a feasible option for obtaining comprehensive clini- EDTA tubes are volume-dependent, and introducing too small a cal pathology data from a single mouse. We also explored how sample into the EDTA tube may result in falsely decreased HCT hydration methods affect the amount of blood that can be drawn and MVC values, falsely increased MCHC, and cause crenation at termination. Often mice are shipped to diagnostic laboratories of RBCs.1,9,10 If the full compliment of blood is placed into the for necropsy and blood samples for clinical pathology evalu- EDTA tube to avoid error, less blood is available for the serum ation are obtained at euthanasia. For shipping, the mice typi- clot tube, thus providing less serum for chemistry analysis. This cally are provided a source of hydration other than water (that study demonstrates that the use of a single tube for generating is, hydration gel). To determine whether the amount of blood

Vol 46, No 2 Journal of the American Association for Laboratory Animal Science March 2007 that could be obtained from a mouse at termination differed, hampered by the small-volume blood samples that can be we gave half of our mice water overnight, and the other half obtained from a single animal. However, the advent of instru- received gel packs as a hydration source. Slightly more blood mentation that requires minute sample volumes and the single- could be obtained from those kept on water overnight than tube method we describe here enhance the utility of clinical those provided gel packs, but the difference was not significant. pathology so that it can be used more extensively in toxicity However, full 14-parameter chemistry profile could be obtained and safety studies as well as biomedical research. more readily from the water group than the gel-pack group. This difference most likely is related to the hydration status of the !CKNOWLEDGMENTS mouse, although this association was not determined specifi- We would like to thank the staffs of the Research Animal Diagnostic cally. We recommend that mice shipped to a diagnostic labora- Laboratory and the Veterinary Clinical Pathology Laboratory at the tory should be housed with water bottles overnight (or longer) University of Missouri for their assistance with this project. prior to necropsy and blood sampling, to increase the likelihood of sufficient sample for clinical pathology evaluation. 2EFERENCES An alternative to shipping mice to a diagnostic necropsy 1. Chen BH, Fong JF, Chiang CH. 1999. Effect of different anticoagu- facility is shipping whole blood. In this study, we placed whole lant, underfilling of blood sample and storage stability on selected blood from several mice in lithium heparin and EDTA tubes and hemogram. Kaohsinung J Med Sci 15:87–93. stored them for various amounts of time before CBC analysis 2. Hall RL. 1992. Clinical pathology for preclinical safety assessment: and preparation of the peripheral blood smear. The duration current global guidelines. Toxicol Pathol 20:472–476. 3. Hall RL. 1999. Clinical pathology: comparative pathophysiology and temperature of blood storage we studied were chosen to and data interpretation. In: Proceedings of American College of simulate various transit times and conditions of shipped blood Laboratory Animal Medicine, Laboratory Animal Medicine: Ap- samples. In both the EDTA and lithium heparin samples, most plication of Clinical Medicine in a Research Environment. San of the CBC parameters were not altered markedly over time, Antonio, TX. consistent with other reports from different species.11 However, 4. Irausquin H. 1992. The value of clinical chemistry data in animal automated platelet counts decreased dramatically after 24 h of screening studies for safety evaluation. Toxicol Pathol 20:515– storage in the lithium heparin but started to increase at 48 and 72 518. 5. Kanno J. 1992. Clinical pathology testing regulatory concerns. h. Despite the increases at 48 and 72 h, platelet counts remained Toxicol Pathol 20:534–537. far below baseline values. A similar pattern of platelet count 6. McClure DE. 1999. Clinical pathology and sample collection in occurred in the EDTA samples, with a less dramatic decrease the laboratory rodent. Vet Clin North Am 2:565–589. at 24 h. In both samples, this pattern is most likely an effect of 7. Medaille C, Briend-Marchal A, Braun JP. 2004. Stability of selected severe platelet clumping at 24 h with some dissolution of the hematology variables in canine blood kept at room temperature clumps at 48 and 72 h. In both the lithium heparin and EDTA in EDTA for 24 and 48 h. Vet Clin Pathol 35:18–23. 8. National Research Council. 1996. Guide for the care and use of tube samples, platelet clumping was readily apparent in the laboratory animals. Washington (DC): National Academy Press. peripheral blood smears at 24 h, with more marked changes in 9. Quimby FW. 1999. The mouse. In: Loeb WF, Quimby FW, editors. the lithium heparin tubes. Similar findings have been reported The clinical chemistry of laboratory animals. Philadelphia: Taylor regarding stored EDTA-treated blood from other species.7 As and Francis. p 3–32. previously stated, WBC decreased and HCT increased over time 10. Stockham SL, Scott MA. 2002. Basic hematologic assays. In: Stock- in the lithium heparin stored blood. This effect is most likely the ham SL, Scott MA, editors. Fundamentals of veterinary clinical result of ruptured WBCs and slight dehydration of the sample pathology. Ames: Iowa State University Press. p 31–48. 11. Vatn S, Framstad T, Torsteinbo WO. 2000. Hematologic evaluation through the storage period. Blood smears from both samples of normal and anemic lambs with the Technicon H*1 using EDTA showed greater number of disrupted WBCs at later time points. or heparin as anticoagulants. Vet Clin Pathol 29:62–68. On the basis of our combined findings, whole blood samples 12. Vignali DA. 2000. Multiplexed particle based flow cytometric stored in either lithium heparin or EDTA likely will provide assays. J Immunol Methods 243:243–255. adequate results after as long as 72 h of storage at 4 pC. How- 13. Wayne PA. 2003. NCCLS document H1-A5, evacuated tubes and ever, differences in automated platelet counts, HCT, and WBC additives for blood specimen collection: approved standard, 5th counts may occur and should be taken into consideration during ed. Wayne (PA): National Committee for Clinical Laboratory Standards. interpretation of diagnostic or research results. 14. Weingard K, Bloom J, Carakostas M, Hall R, Helfrich M, Latimer As part of a hematology evaluation in mice, a bone marrow K, Levine B, Neptun D, Rebar A, Stitzel K, Troup C. 1992. Clinical 2,15 smear is recommended. Bone marrow samples for a smear are pathology testing recommendations for nonclinical toxicity and obtained easily from the mouse femur at necropsy, and micro- safety studies. Toxicol Pathol 20:539–543. scopic evaluation of these smears is usually reserved for when 15. Weingard K, Brown G, Hall R, Davies D, Gossett K, Neptun D, abnormalities are noted in the CBC or blood smear. Regardless Waner T, Matsuzawa T, Salemink P, Froelke W, Provost JP, Dal of differences regarding which should be examined,2,15 bone Negro G, Batchelor J, Nomura M, Groetsch H, Boink A, Kimball J, Woodman D, York M, Fabianson-Johnson E, Lupart M, Mel- marrow smears should be obtained as part of routine clinical loni E. 1996. Harmonization of animal clinical pathology testing pathology sample collection. in toxicity and safety studies. Fundam Appl Toxicol 29:198–201. Historically, the use of clinical pathology in mice has been