The amount of blood you can safely withdraw depends on the size of the mouse, how frequently you sample its blood, and whether you give it fluids afterwards. Below are some guidelines for single samples, multiple samples, and exsanguination.
Single sample
Most animals go into shock if 25-30% of their blood volume is rapidly removed; over 50% die if 30-40% is removed; and nearly all die if more than 40% is removed (Walcott 1945; Smith and Hamlin 1977; Waynforth and Flecknell 1992; Kristensen and Feldman 1995).
For most species, the blood volume in mililiters is approximately 6-8% of their body weight (BW) in gm, or 55-80 ml/kg BW. However, this relationship varies significantly among species. In general, without fluid replacement, approximately 10% of the total blood volume (0.75% of BW) can be safely removed at one time; with fluid replacement, up to 15% (approximately 1.5% of BW) can be removed (Smith and Hamlin 1977; McGuill and Rowan 1989; Authement 1992; Waynforth and Flecknell 1992).
The average total blood volume of a mouse is about 77-80 ml/kg (0.077-0.080 ml/gm) (Mitruka and Rawnsley 1981; Harkness and Wagner 1989). Without fluid replacement, approximately 0.007-0.008 ml of blood/gm BW can be safely withdrawn (about 0.18-0.2 ml of blood from a 25-gm mouse); with fluid replacement, approximately 0.014-0.016 ml blood/gm BW can be withdrawn (about 0.35-0.4 ml blood from a 25-gm mouse). Generally, replacement fluids are given subcutaneously and should be warmed beforehand.
Multiple samples
If multiple samples must be taken at short intervals, smaller volumes should be removed each time. The issue is not fluid loss: although blood volume is rapidly replaced following acute blood loss (Fettman 1985), other blood components may not be restored for several weeks (Jain 1986; McGuill and Rowan 1989; Waynforth and Flecknell 1992). For this reason, the quantity of blood withdrawn per week should not exceed 7.5% of the total blood volume, or 10% of the total blood volume every two weeks (McGuill and Rowan 1989).
Ideally, if a mouse will have large amounts (i.e., approaching the recommended maximum) of blood withdrawn more than three times at intervals of two weeks or less, its hematocrit and/or hemoglobin should be monitored. If deficits are found in these parameters, sampling volume and/or frequency should be reduced.
If samples are withdrawn weekly, approximately 0.006 ml of blood per gm BW can be taken; if withdrawn every other week, approximately 0.008 ml blood/gm BW can be taken. For a 25-gm mouse this corresponds to 0.15 ml per week or 0.2 ml every two weeks. For repeated sampling at intervals of three weeks or less, fluid replacement will not allow safe removal of larger volumes at a time.
To maintain the optimal health and physiological stability of a mouse, blood volume withdrawn (whether repeatedly or only once) should be limited to the lower end of the suggested range. Higher-end volumes should be withdrawn only if necessary and only from mice in peak health. The amount of blood that can be safely removed from sick, severely stressed or otherwise abnormal mice is less (perhaps even below the suggested ranges) than that which can be taken from a healthy, minimally stressed mouse.
How much can I take as a terminal procedure?
Under ideal conditions, a skilled person can obtain approximately 50-75% of a mouse's total blood volume (3-4% of BW) by exsanguination. This amounts to 0.04-0.06 ml blood/gm BW, or 1.0-1.5 ml blood from a 25 gm mouse. The best yields are obtained if the blood is removed slowly and steadily so that the heart is kept beating as long as possible.
The total volume of blood cells obtained can be increased by giving the mouse fluids during bleeding to maintain its blood pressure. This is best accomplished by using a vascular catheter to alternate between blood withdrawal and fluid administration. (Note: Exsanguination is distressful and should be performed only on anesthetized mice.)
References
Authement JM. 1992. Blood transfusion therapy. In: DiBartola SP, editor. Fluid therapy in small animal practice. Philadelphia: W.B. Saunders Company; 720 p.
Fettman MJ. 1985. Hypertonic crystalloid solutions for treating hemorrhagic shock. Compend Contin Educ Pract Vet 7:915-20.
Harkness JE, Wagner JE. 1989. Biology and husbandry. In: Harkness JE, Wagner JE, editors. The biology and medicine of rabbits and rodents, 3rd ed. Philadelphia: Lea & Febiger; 372 p.
Jain NC. 1986. Schalm's Veterinary Hematology. Philadelphia: Lea & Febiger; 1221 p.
Kristensen AT, Feldman BF. 1995. Blood banking and transfusion medicine. In: Ettinger SJ, Feldman EC, editors. Textbook of veterinary internal medicine, 4th ed. Philadelphia: W.B. Saunders Company; 2145 p.
McGuill MW, Rowan AN. 1989. Biological effects of blood loss: implications for sampling volumes and techniques. ILAR News 31:5-20
Mitruka BM, Rawnsley HM. 1981. Clinical, biochemical and hematological reference values in normal experimental animals and normal humans. New York: Masson Publishing; 413 p.
Smith CR, Hamlin RL. 1977. Circulatory shock. In: Swenson MJ, editor. Dukes' physiology of domestic animals. 9th ed. Ithaca, NY: Cornell University Press; 914 p.
Walcott WW. 1945. Blood volume in experimental hemorrhagic shock. Am J Physiol 143:247.
Waynforth HB, Flecknell PA. 1992. Anesthesia and postoperative care. In: Waynforth HB, Flecknell PA, editors. Experimental and surgical technique in the rat, 2nd ed. LondonL Academic Press; 382 p.