Comparison of Novel and Traditional Bleeding Techniques in Neonatal and Juvenile Mice
Blood collection is frequently used for neonatal and juvenile mice in toxicology, developmental, and immunology studies and is often a terminal procedure. However, the use of nonterminal blood collection techniques, including the submandibular and the submental collection techniques described for adult mice, may offer opportunities to reduce animal numbers and refine current methods. The use of the submental technique has not been described for neonatal or juvenile mice. In this study, we compared the submental and submandibular blood collection techniques to determine their suitability for use in neonatal and juvenile mice. Male and female CD1 mice, ages 7, 14, 21, and 28 d, were randomized by sex into submental (n = 16), submandibular (n = 16), or control (n = 8) groups. Each mouse was weighed, bled per its assigned group (or only restrained in the case of control mice), and then decapitated without anesthesia for terminal blood collection. Blood collection volume and corticosterone concentrations were measured. The 2 methods showed significant differences in the volume of blood collected at ages 14 and 28, with the submandibular technique yielding significantly higher volumes. No significant differences were detected in corticosterone levels between the 2 techniques based on age or sex. A subset of mice (n = 8, 2 per age group) were bled via submental or submandibular technique and were evaluated 48 h later for gross and histopathologic evidence of trauma. Seven of the 8 mice showed expected inflammation and healing at the collection sites, with 4 mice having embedded strands of fur in the tissue. These data indicate that the submental blood collection is a viable method for nonterminal blood collection method in neonatal and juvenile mice, especially when smaller amounts of blood are needed.Abstract
Introduction
Mice and rats comprise 95% of the animals used in biomedical research.20 Due to their anatomic, physiologic, and genetic similarities to humans,3 these rodents are integral to drug discovery, model development, and studying disease processes. Most studies use adult mice; these data may then be extrapolated and applied to younger mice. However, neonatal mice may respond differently when compared with adult mice (for example, neonates have higher sensitivity to genotoxic mouse carcinogens).10,12 These differences are due largely to changes in physiologic parameters and developmental differences during the early stages of growth, as has been reported in multiple species.21,28,35 Blood composition differs in newborn and adult mice.10,31,36 Therefore, the use of neonatal and juvenile mice, as compared with the extrapolation of data from adult mice, is essential to the generation of valid data relevant to these younger animals.
Neonatal mice have been used to study developmental biology, immunology, adeno-associated viral therapeutics, reproductive toxicology, gut microbiology, and tumorgenicity.10,14,27,28 Evaluation of blood values is an informative parameter for such studies. Blood collection from neonates often uses terminal techniques such as decapitation or cardiac puncture.27 These terminal methods require the use of a larger number of mice, especially when neonates must be sampled at various time points. The 3 Rs, reduction, refinement, and replacement, should always be considered when performing animal research.4 Our objective here is to assess 2 methods of survival blood collection to reduce animal numbers required for studies, to refine blood collection techniques, and thereby to improve animal welfare and adhere to the principles of the 3 Rs.
The use of survival blood collection instead of terminal collection in young mice could allow the collection of longitudinal data from individual mice over time. The submandibular blood collection technique was first described in 2005, is in common use, and involves blood collection from the facial veins of mice.16 However, some studies that used this technique have reported unexpected complications such as excessive hemorrhage, damage to nerves and facial muscles, and lethargy.11,23,33 The original study did not evaluate submandibular blood collection using neonatal or juvenile mice, and no studies have evaluated this method in young mice to our knowledge. However, one study has used submandibular sampling, also referred to as facial vein sampling, for survival blood collection in mice of varying ages, including neonates and juveniles.7
The submental blood collection technique was described in 201629 and demonstrated blood collection from the inferior labial vein, a branch of the facial vein.8 Although recent papers have further evaluated this technique,1,13,29 none have evaluated the submental method in neonatal and juvenile mice. Comparing blood collection techniques in neonatal and juvenile mice with regard to welfare, ease of collection, and sample integrity would determine whether this method is as effective as submandibular bleeding, indicate a possible refinement, and suggest a more accessible and reliable method for blood collection from young mice.
This study compared the submandibular and submental blood collection techniques in neonatal and juvenile mice. Measures evaluated were stress, welfare, and sample quality. We hypothesized that the submental blood collection method would be less stressful for the mice (yielding lower levels of blood glucose and serum corticosterone), faster to perform (shorter time for collection), and more efficient (fewer attempts to successfully collect blood) as compared with the submandibular blood collection method. We also hypothesized that the samples collected from the 2 methods would be comparable in quality (amount of blood collected, degree of platelet clumping, and lysed white blood cells).
Materials and Methods
Animals.
A total of 228 Crl:CD1(ICR) mice (Charles River Labs, Morrisville, NC) were used. All mice were housed in an AAALAC-accredited vivarium, and care was provided in accordance with the Guide for the Care and Use of Laboratory Animals.24 All animal protocols were approved by the University of North Carolina at Chapel Hill IACUC.
Mice were housed in nonautoclaved, individually ventilated polysulfone cages (Green line, Sealsafe Plus GM500, and DGM Racks; Tecniplast, West Chester, PA) that were processed through a tunnel washer reaching at least 82 °C (180 °F) and changed biweekly. The cages contained corncob bedding (1/4 in., irradiated; Bed-O’ Cobs; Lab Supply, Durham, NC) with a cotton square for nesting material (Ancare, Bellmore, NY) and red transparent hut for enrichment (Mouse Hut; Bio-Serv, Flemington, NJ). Mice received pelleted feed (5V5R PicoLab Select Rodent 50 IF/6F; LabDiet, Durham, NC) ad libitum and reverse-osmosis-purified water via an automated water system (Edstrom Automated Watering Systems; Avidity Biosciences, San Diego, CA). The room was maintained on a 12:12-h light:dark cycle (lights on 0700 and lights off 1900). Mice were weaned at approximately 24 d of age into cages with no more than 5 mice per cage, separated by sex. Rooms were kept 22 ± −16 °C (72 ± 2 °F) and 30 to 70% humidity. Colony health status was monitored by using live sentinel mice that were exposed to soiled bedding and tested 3 times yearly using multiplex fluorescent immunoassay v2 and PCR (IDEXX BioAnalytics, North Grafton, MA). Mice were free of pinworms (Syphacia muris, S. obvelata, and Aspiculuris tetraptera), fur mites (Mycoptes, Radfordia, and Myobia spp.), epizootic diarrhea of infant mice, murine hepatitis virus, mouse parvovirus, minute virus of mice, Theiler murine encephalomyelitis virus/GDVII, ectromelia virus, lymphocytic choriomeningitis virus, mouse adenovirus type 1/2, mouse cytomegalovirus 1, polyoma virus, pneumonia virus of mice, reovirus 3, Sendai virus, Mycoplasma pulmonis, and Filobacterium rodentium (cilia-associated respiratory bacillus).
Pilot study.
A single veterinarian who had been trained and was experienced in both submandibular and submental blood collection techniques in rodents but does not regularly perform them in daily practice carried out all procedures described in both the pilot study and the main study. The pilot study used mice to optimize handling, restraint, and collection techniques for the main study. The main study results included no data from mice used in the pilot study. Two breeding pairs each produced 2 litters of mice. Five mice per age group (7, 14, 21, and 28 d of age) were sampled by using either submental or submandibular approach. Sex was not evenly represented but was accounted for in the main study. For 7-, 14-, and 21-d-old mice, each mouse was removed from the dam, sexed, weighed, and placed on the wire food hopper of a cage. Submental or submandibular blood collection was performed using a 3-mm Goldenrod lancet (Medipoint, Mineola, NY) while the mouse was restrained by hand. After initial trials, a 4-mm lancet was used on the remaining mice. Blood was collected from the site using a heparinized microhematocrit capillary tube (Thermo Fisher Scientific, Waltham, MA). No more than 2 skin puncture attempts were allowed per collection. After blood was collected in the heparinized microhematocrit capillary tube, the mouse was released from restraint and monitored for clinical signs and cessation of bleeding. The time between the initiation of restraint and cessation of bleeding was recorded. The mouse then underwent decapitation without anesthesia. Trunk blood was collected in an anticoagulant-free serum separator tube (Becton, Dickinson and Company, Franklin Lakes, NJ). Figure 1A and B show the group distributions and workflow for the study.
Citation: Journal of the American Association for Laboratory Animal Science 63, 3; 10.30802/AALAS-JAALAS-23-000116
Experimental design.
All techniques were performed by the same individual as above. Three experimental groups were evaluated at 4 different age groups (7-, 14-, 21-, and 28-d-old): the 3 groups were submandibular (n = 16), submental (n = 16), and control mice (n = 8). These numbers were determined by a statistician to provide 80% power for 2-sided tests with 5% levels of significance. Each group had equal numbers of male and female mice. Before blood collection, neonatal mice (defined as 0- to 14-d-old) were removed from the dam and placed in a clean, heated cage. Juvenile mice (defined as 15- to 28-d-old) were removed from the dam and placed in a clean cage without heat. At their specified ages, mice were sexed, weighed, and assigned to an experimental group based on sex and age by using an Excel (Microsoft, Redmond, WA) randomization tool that created a consecutive number list corresponding to the experimental group. For all groups, conscious, unanesthetized mice were placed on a wire feeder in a cage for restraint to replicate a real-life scenario of how and where the technique would be performed. Once appropriate landmarks were identified (as described below), a 4-mm lancet (Goldenrod; Medipoint, Mineola, NY) was used to pierce the tissue, and blood was collected with a heparinized microhematocrit capillary tube as described in the pilot study. No more than 2 attempts were allowed per mouse, and the number of attempts was recorded. After blood collection, the mouse was released onto a clean surface, where it was monitored until recovered, as evidenced by a lack of lethargy, weakness, or continued bleeding. The time interval from restraint to recovery was recorded. Mice were monitored during blood collection for vocalization, defecation, urination, or any abnormal clinical signs. Once these data were collected, mice were manually restrained and underwent unanesthetized decapitation. Trunk blood was collected into a yellow serum separator tube (Becton, Dickinson and Company, Franklin Lakes, NJ) and placed on ice, and a glucose measurement (Contour next EZ, Blood Glucose Meter, 9628; Ascensia Diabetes Care, Parsippany, NJ) was recorded. Control mice were placed on top of the wire feeder and manually restrained for approximately 35 s (based on average time for blood collection found in the pilot study). They were then released, restrained again, and decapitated without anesthesia; blood was collected into a yellow serum separator tube that was placed on ice. Blood glucose was then measured. All trunk blood samples were centrifuged (Mini Centrifuge; Labnet International; C1301-CEP) within 4 h of collection, and the serum was separated into microtubes and placed in a −80 °C freezer for future evaluation. All data collection took place roughly at the same time of day in the afternoons during a 4-h window. See Figure 1C and 1D for study group distribution and workflow.
Eight additional, randomly selected mice that underwent submandibular or submental blood collection (n = 2/age group) were returned to the dam and the home cage after blood collection. At 48 h after blood collection, they were euthanized by carbon dioxide inhalation and cervical dislocation for gross and histologic evaluation of the collection site. The data from these mice were not used in the analysis of the main study described above.
Submandibular blood collection.
The mouse was restrained by gently grasping the skin of the neck using the one-handed “scruffing” method, as previously described.32 A 4-mm lancet was used to puncture the cheek caudal dorsal to the whorl of hair along the mandible. The technique was performed as outlined in the literature.16 The landmark (red circle) and area to puncture (red x) are visualized in Figure 2A.
Citation: Journal of the American Association for Laboratory Animal Science 63, 3; 10.30802/AALAS-JAALAS-23-000116
Submental blood collection.
For the pilot study, submental blood collection was performed as previously described.27 Briefly, the mouse was restrained by immobilizing and tilting the head back to visualize the ventral jaw and neck. A group of hairs on the midline of the neck served as a starting location, with the areas to puncture located slightly rostrolateral on either side of these hairs.29 During our pilot, several mice bled more than necessary from those locations. After consultation with an individual who regularly uses this technique (N. Riddick, personal communication, December 22, 2022), we modified our technique puncturing a few millimeters rostral to the group of hairs along the midline; this site yielded sufficient blood volume, was more consistent in successful puncture, and had slower blood flow, reducing blood loss. Therefore, the main study was conducted using this modified technique, shown in Figure 2B. Although we used this modified submental technique in this study, we will refer to it as “submental” in the remainder of the paper for ease of reading.
Corticosterone measurement.
Stored serum samples were shipped overnight with dry ice to a diagnostic laboratory (Debra L. Hickman, Clinical Professor, Department of Comparative Pathobiology, Purdue University, West Lafayette, IN) for measurement of corticosterone. Serum corticosterone was measured using a commercially available corticosterone ELISA kit (K014-H1/H5; Arbor Assays, Ann Arbor, MI). Serum samples were not diluted. The plates were read on an ELISA plate reader set to 450 nm using SoftMax Pro 7.0 (Molecular Devices, Sunnyvale, CA). Concentrations were calculated using the 4-parameter logistic curve assay with an online data analysis tool (http://www.myassays.com/four-parameter-logistic-curve.assay).
Sample volume, quality, and glucose evaluation.
The length of the collection tube that contained blood was measured in millimeters to quantify the blood volume collected. Blood glucose was measured from tube and trunk blood using a glucometer (Contour next EZ, Blood Glucose Meter, 9628; Ascensia Diabetes Care, Parsippany, NJ). A drop of blood from the tube was placed on a glass slide and a blood film was created. The slides were stained with a commercially available modified Romanowsky stain (Epredia Shandon Kwik-Diff Stain, Fisher Scientific, Pittsburgh, PA), and the feathered edge was evaluated for platelet clumps and lysed white blood cells. The first 10 white blood cells observed, starting at the feather edge, were scored as being lysed or not to determine the presence of lysis.
Histology.
Mouse heads were fixed in 10% neutral buffered formalin for 1 to 2 wk before being transferred to 70% ethanol. The skin was reflected at each procedure site, and the procedure site was cut into longitudinal strips. The underlying muscle was placed in the same cassette and trimmed in the same orientation. Tissues were routinely processed and embedded with paraffin wax using an automated tissue processor (Leica Biosystems). The paraffin-embedded tissue blocks were then sectioned into 5-μm-thick slices and placed on positively charged slides. The slides were stained with hematoxylin and eosin and then coverslipped. A boarded veterinary pathologist examined the slides for signs of inflammation and other features of tissue damage. Representative images were captured using a Lumenera INFINITY5 digital camera (Teledyne Lumenera, Ottawa, ON, Canada) mounted on an Olympus BX53 microscope (Evident Scientific).
Statistical analysis.
A 2-sided P value of less than or equal to 0.05 was considered statistically significant for all statistical analyses. A linear regression model was used to analyze all quantitative data, including corticosterone, blood glucose from the blood collection site, serum glucose collected from the trunk, duration of the procedure, and blood volume collected. All quantitative data were evaluated, with the models accounting for age, gender, and blood collection technique with all 3 interactions (3-way ANOVA). Binary outcomes such as urination, defecation, vocalization, number of attempts, clotting, and lysed white blood cells were analyzed using a logistic regression. Only main effects were considered because of zero values for some binary parameters. Glucose values in whole blood were compared between the collection site and the trunk by using repeated measures of regression. All statistical methods were performed using SAS 9.4 (SAS Institute, Cary, NC).
Results
Time to collection and number of attempts.
Blood collection required significantly more time for the submental technique on day 7 (P = 0.0481). However, the 2 sites were not different on days 14 (P = 0.9867), 21 (P = 0.0506), or 28 (P = 0.1107) (Figure 3A). The number of attempts needed for each technique and across age groups was not significantly different.
Citation: Journal of the American Association for Laboratory Animal Science 63, 3; 10.30802/AALAS-JAALAS-23-000116
Amount of blood collected.
The submental technique yielded a significantly lower blood volume than did the submandibular technique when data for all ages were analyzed together (P < 0.0001, Figure 3B). The amount of blood collected from the 2 sites was not significantly different on days 7 or 21 but was significantly lower on days 14 (P = 0.0038) and 28 (P = 0.0056).
Serum corticosterone.
Serum corticosterone concentrations were not significantly different between the 2 methods within individual age groups, ranging between 817 and 1,787 pg/mL (Figure 4A). For day 7, blood corticosterone could not be measured in 18 of the 40 mice (9 submental, 7 submandibular, and 3 control) due to insufficient sample volume. Values from sampled mice were not significantly different from those of control mice.
Citation: Journal of the American Association for Laboratory Animal Science 63, 3; 10.30802/AALAS-JAALAS-23-000116
Glucose measurements.
Glucose was significantly higher in the submental group (average of 162 mg/dL) as compared with the submandibular group (average, 149 mg/dL) on day 28 (P = 0.0387) (Figure 4B). Otherwise, glucose levels were not different in blood collected from either in vivo site or from the trunks of all study and control mice (Figure 4C). When comparing in vivo and trunk blood values from the same mouse, values from the collection site were significantly lower than those measured in trunk blood in some cases (P = 0.0354; Table S1).
Blood smears.
No significant differences in platelet clumps or lysed white blood cells were detected when comparing blood collection techniques among the age groups.
Clinical and gross pathology observations.
Observations of vocalization, defecation, and urination showed more significant differences with respect to age rather than collection method. Overall, mice bled by the submandibular technique were significantly less likely to defecate during the procedure as compared with control and submental mice (P < 0.01). Mice bled by the submental technique vocalized significantly less than control mice (P = 0.0463) but were not different from the submandibular group (P = 0.7277). Day 28 mice vocalized and urinated more than day 7 and day 14 mice (P < 0.05) and defecated more than day 7 mice (P < 0.05). Day 21 mice urinated more than day 14 mice (P = 0.0383), and day 21 and day 14 mice defecated more than day 7 mice (P < 0.05) (Tables S2–S4). Of the 128 mice bled in both groups, one day 28 mouse and 3 d 14 mice in the submandibular group had blood in the ear canal ipsilateral to the puncture. No such observations were seen in the submental group. Study groups showed no significant difference in the weight within the 4 age groups. None of the parameters measured showed significant differences between males and females.
Gross necropsy of randomly selected mice from each age and treatment group revealed that mice in the submandibular group had a focal area of hemorrhage in the muscle and adjacent subcutis on days 14 (Figure 5H), 21 (Figure 5L), and 28 (Figure 5P), corresponding with the approximate puncture location. A small, flat, pinpoint area of hemorrhage (< 1 mm) was seen at the epidermal submental puncture site on a single day 14 mouse (Figure 5E, black arrow). No other mice had visible lesions on the skin surface or subcutis (Figure 5A, B, C, D, F, G, I, J, K, M, N, and O).
Citation: Journal of the American Association for Laboratory Animal Science 63, 3; 10.30802/AALAS-JAALAS-23-000116
Seven of the 8 mice that were euthanized 48 h after blood collection had expected inflammation in all age groups for both techniques (Figure 6). Inflammation was not detected for the day 21 submandibular mouse (Figure 6K and L) and on two muscle and one skin sample (Figure 6H, M, and O). A low number of macrophages, neutrophils, and lymphocytes were present in the muscle and subcutaneous and perivascular spaces together with mild myofiber degeneration and regeneration (Figure 6A-G, I, J, and N). Extravasated red blood cells were sometimes observed in macrophages (Figure 6P, black arrow), indicating hemorrhage. Of the 7 mice with signs of inflammation, 4 showed hairs embedded in the muscle, surrounded by abundant viable neutrophils, fewer macrophages, and rare lymphocytes (Figure 6E, G, I, and J). The hairs were observed on day 14 for both techniques and on days 21 and 28 for the submental site.
Citation: Journal of the American Association for Laboratory Animal Science 63, 3; 10.30802/AALAS-JAALAS-23-000116
Discussion
Our pilot study addressed potential adverse effects and allowed us to modify our submental technique for use in young mice. A previous report on the submental technique described a target area 2 mm rostrolateral to midline.29 However, our application used a modified technique, described above, that yielded consistent results. The pilot revealed that the 3-mm lancets were often too small to allow a large enough puncture hole for blood to escape from the skin, as evidenced by subcutaneous pooled blood with none exiting the puncture site. The submandibular site also generally produced enough blood to fill the 75 μL heparinized microhematocrit capillary tube; this could result in overcollection of blood, potentially leading to anemia or death. Therefore, we modified our approach by filling the capillary tube to only one-third of its length, approximately 25 μL, for both collection sites.
Our main study demonstrated that the submental and submandibular techniques were comparable with regard to stress, time for collection, and sample quality in neonatal and juvenile mice. The most significant difference was the amount of blood collected from each site; a significantly larger blood volume was collected from the submandibular site. This could be due to the underlying anatomy of the blood vessels being targeted with each technique. As previously described, the submental vein runs obliquely toward the chin of the mouse.8 This vessel is described as being too small to create a drip rate necessary for blood collection, yet it is the target vessel for our modified submental technique. The area targeted by previous papers1,13,29 was most likely the inferior labial vein or its convergence with the facial vein; this site could allow collection of larger blood volumes in neonatal and juvenile mice, but additional studies would be necessary to test this. The targeted vessel for the submandibular method is the facial vein, which branches from the external jugular vein and often yields an adequate drip rate for blood collection.16 Knowledge regarding the vein anatomy and flow rate of this vein can help to guide researchers toward the optimal blood collection technique for their purposes. The submental site may be preferable if only a small volume is required. Blood collection volume is important in neonatal and juvenile mice due to the risk of overcollection, which can lead to anemia or death. The submental site may provide a more controlled method for collecting small blood quantities. The submandibular technique may be better if more blood is required.
Corticosterone is often used as a measure of stress in mice.17 The secretion of corticosteroids increases when the hypothalamus-pituitary-adrenal axis is activated in response to stress, resulting in a rise in serum corticosterone levels.19 We found no differences in serum corticosterone levels among submental, submandibular, and control groups regardless of age. This finding suggests that stress was similar for all groups of mice, including controls. This similarity suggests that the stress experienced through handling or euthanasia contributed more to the increases in corticosterone than did the in vivo collection techniques. Thus, the submental technique does not appear to be significantly more stressful than the submandibular technique.
Blood glucose values are common biomarkers for stress in rodents.15 Rodent blood glucose levels are directly affected by the hypothalamic-pituitary-adrenal axis.15 On day 28, mice in the submental group had significantly higher blood glucose levels than did mice in the submandibular group. However, blood glucose values in the other age groups do not indicate that mice in the 3 groups experienced differences in stress. Comparison of in vivo and trunk blood glucose measurements allowed us to examine whether methods of blood collection affect data. Previous studies reported varying levels of glucose and insulin depending on the collection site in conscious mice.5,6,9,30 In our study, glucose levels measured in trunk blood were significantly higher than those measured for days 7 submental, 21 submental, and 28 submandibular mice, suggesting the blood collection site affects glucose values. Further studies could evaluate differences in glucose values among different collection sites. We used a point-of-care glucometer that is commonly used in rodent research; however, one study reported that tabletop glucometers generate higher blood glucose levels than biochemical tests.34 Because our study compared the differences among groups, we found that the glucometer gave consistent and reproducible results.
Other than day 7, the time needed to perform blood collection time was comparable between the 2 techniques for all groups. On day 7, blood collection from the submandibular and submental routes required approximately 26 and 33 s, respectively. However, a 7-s difference is not operationally significant. Thus, the sites are both equivalent and practical from a time perspective.
We compared the frequency of urination, defecation, and vocalization during blood collection between blood collection and control groups as an indirect measure of stress. In general, mice from the submandibular group defecated less and the submental group vocalized less as compared with control mice, but the submandibular and submental groups were similar. We did not control for food intake but assumed similar variation in food consumption among individuals of the same age for both study and control groups. Because we did not control food intake, variations in defecation and urination were likely, and the difference in diet (milk from the dam or lab diet when older) could also have influenced these 2 variables. Mice that were 7 and 14 d old vocalized significantly less than did 28-d-old mice. Perhaps the younger mice simply vocalized less audibly than the older mice, but we did not measure the ultrasonic vocalizations that mice emit during events such as separation from the dam.18 Future studies could analyze ultrasonic vocalizations of neonatal mice as a measure of stress. Overall, urination, defecation, and vocalization were not significantly different between the groups, suggesting that the 2 techniques were associated with similar levels of outward stress.
Trauma to the cheek area is a known adverse effect of submandibular blood collection from mice.11,23,33 In our study, 4 mice in the submandibular had fresh blood in the ear canal ipsilateral to the collection site. The target area is closely associated with the ear canal and oropharynx, and a network of blood vessels and nerves is present in the peripheral region of puncture.11 Younger mice may have a greater risk of these adverse effects due to their small size. Our data suggests that the submental method reduces risk of damaging closely associated structures in younger mice. The tongue, which lies dorsal to the blood collection site, is a structure of concern, but no clinical signs, such as gross blood in the mouth, were observed during this study or were reported in other studies evaluating this technique.1,13,29 Although the submental bleeding technique may cause less peripheral damage than the submandibular technique, further studies using a subjective clinical scoring system and a grimace scale could further evaluate the clinical impact of these techniques.
No significant differences in clinical signs or gross or microscopic lesions were observed among technique or age groups. For both techniques, small scabs were observed grossly on day 14 and 28 mice. Microscopically, the anticipated levels of inflammation, myofiber degeneration and regeneration, and red blood cell (hemorrhage) scavenging by macrophages were similar across all techniques and age groups. Submandibular histopathology from 21- and 28-d old mice showed less inflammation than was seen in the younger mice, indicating that submandibular collection was slightly less traumatic in older mice. Additional research would be necessary to confirm and explain this finding.
Our study had several limitations. First, one individual performed all experimental manipulations to reduce variation and provide a consistent comparison between study groups; however, an individual could be biased for one of the techniques due to preference or skill level. Second, mice were not fasted before blood collection, such that the time of consumption or amount of food ingestion could influence glucose levels. Fasting overnight is often used in advance of performing metabolic tolerance tests.2 Although such studies compare the effects of fasting duration and time of day, the most critical component is consistency.2,25,26 Fasting neonatal and juvenile mice are more challenging than adults, as their small size and high metabolic rate may lead to hypoglycemia during fasting.22 Therefore, we opted to forego fasting for measuring glucose.
In conclusion, our study indicates that our modified submental blood collection technique is equivalent to the submandibular site with regard to stress for neonatal and juvenile mice and may be a better option if small amounts of blood are needed. Our study design focused on comparative measures of stress, consistency between sampling sites, and microscopic features of the sampling site. Study design and individual skill should be considered in choosing between these sites.
Supplementary Materials
Table S1.Glucose from bleed site compared to glucose from trunk across age and technique. *, P ≤ 0.05; †, P ≤ 0.01.
Table S2.Urination instances compared across sex, group, and age. *, P ≤ 0.05; †, P ≤ 0.01.
Table S3.Defecation instances compared across sex, group, and age. *, P ≤ 0.05; †, P ≤ 0.01.
Table S4.Vocalization instances compared across sex, group, and age. *, P ≤ 0.05; †, P ≤ 0.01.
(A) Design of pilot study. (B) Workflow of pilot study. The blue box indicates the period during mice were observed for adverse clinical observations and the time needed for blood collection was measured. (C) Design of main study. (D) Workflow of main study. The blue box indicates the period during which mice were observed for adverse clinical observations and the time used for blood collection was measured.
Visualization of blood collection location in 7-d-old Crl:CD1(ICR) mouse for the (A) submandibular technique, and (B) modified submental technique. Circle denotes the anatomic landmarks; X denotes the puncture site.
Clinical assessment of blood collection techniques. Circles (blue) indicate individual mice in the submental group. Triangles (red) indicate individual mice in the submandibular group. (A) Time needed to collect blood. Groups were significantly different on day 7. (B) Volume of blood collected. Groups were significantly different on days 14 and 28. *, P ≤ 0.05.
Evaluation of blood parameters. Circles (blue) indicate individual mice in the submental group. Triangles (red) indicate individual mice in the submandibular group. Squares (green) indicate individual control mice. (A) Serum corticosterone. No significant differences. (B) Whole blood glucose from collection site. *, Significant differences between day 28 submental and submandibular sites (P ≤ 0.05). (C) Whole blood glucose values were not significantly different in submental and submandibular groups. ^, Significant differences between technique (B) and trunk blood (C) within age group (P ≤ 0.05).
Representative views of gross pathology at 48 h after blood collection in different age groups. (A, E, I, M) Submental group epidermal surface. (B, F, J, N) Submental group after skin removal. (C, G, K, O) Submandibular group epidermal surface. (D, H, L, P) Submandibular group after skin removal. (A, B, C, D) Day 7. (E, F, G, H) Day 14. (I, J, K, L) Day 21. (M, N, O, P) Day 28. (A, B, C, D, F, G, I, J, K, M, N, O) No lesions present. (E) Arrow indicates small pinpoint lesion. (H) Arrow indicates area of subcutaneous hemorrhage. Arrowhead shows hemorrhage in muscle. (L) Arrow indicates area of subcutaneous hemorrhage. Arrowhead shows hemorrhage in muscle. (P) Arrow indicates area of subcutaneous hemorrhage. Arrowhead shows hemorrhage within muscle.
Representative images of microscopic changes at 48 h after blood collection in different age groups. (A) Submental muscle. Arrow indicates muscle damage and inflammation consisting of a few neutrophils, macrophages, and lymphocytes. (B) Submental skin with a small focus of inflammation at the edge of the deep dermis indicated by the arrow. (C) Submandibular muscle. Arrow indicates inflammation. (D) Submandibular skin. Arrow indicates serocellular crust on the epidermal layer that is superficial to a region of inflammation that extends into the underlying subcutis. There is mild muscle damage and inflammation at the deep edge of the sample. (E) Submental muscle. Arrow indicates a piece of hair embedded in musculature surrounded by inflammatory cells, and mild muscle damage. (F) Submental skin. Arrow indicates serocellular crust on the epidermal layer. (G) Submandibular muscle. Arrow indicates a hair shaft embedded in muscle surrounded by inflammatory cells (H) Submandibular skin with no significant findings. (I) Submental muscle with hair shaft. Arrow indicates a piece of hair embedded in musculature surrounded by abundant inflammatory cells and some moderate, localized muscle damage. (J) Submental muscle. Arrow indicates a piece of hair surrounded by hemorrhage, myocyte degeneration and necrosis, and mild inflammation. (K) Submandibular muscle with no significant findings. (L) Submandibular skin with no significant findings. (M) Submental muscle. No significant findings. (N) Submental skin with a mild increase in inflammatory cells within the subcutis. (O) Submandibular muscle with no significant findings. (P) Submandibular muscle with extravasated red blood cells (hemorrhage), some inside macrophages (hemophagocytosis) indicated by the arrow. This suggests the initial stages of wound healing. Slides were stained with hematoxylin and eosin. (A, B, C, D) Day 7. (E, F, G, H) Day 14. (I, J, K, L) Day 21. (M, N, O, P) Day 28. Original magnification: 100× (C, D, E, H, I, K, L, M, O), 200× (A, B, F, G, J, N), and 400× (P).
Contributor Notes
This article contains supplemental materials online.