Laboratory Animals (1988) 22, 177-182 177 An improved simple technique for the collection of blood samples from rats and mice G. CONYBEARE, G. B. LESLIE, K. ANGLES, R. J. BARRETT, J. S. H. LUKE & D. R. GASK Department of Toxicology, Smith Kline & French Research Ltd, The Fry the, Welwyn, Herts AL6 9AR, UK Summary The technique of blood collection from the lateral caudal vein has been improved. The method requires only moderate skill and no anaesthesia is necessary. Collection of blood samples causes little trauma and can be repeated at frequent (8 h) intervals thus making the method particularly suitable for pharmacokinetic and hormonal studies. Blood samples are uncontaminated by tissue fluids. Large volumes can be obtained, in mice up to 1 5 ml and in Wistar rats over 4 mi. Keywords: Blood collection; Rat; Lateral tail vein; Venipuncture Methods for intravenous injection and collection of venous blood from rats have been extensively reviewed by Cocchetto and Bjornsson (1983). In toxicology, drug metabolism and pharmacokinetic studies those blood collection methods which require anaesthesia or surgery are often inappropriate because of possible interference with assay methods or with normal phys~ology. Many methods cause considerable trauma and do not lend themselves to frequent repetition or do not permit the collection of large volumes of blood. Some methods require considerable skill or can put the lives of the animals at risk which would be most undesirable in long term toxicity studies. Collection from the dorsal metatarsal vein has been described by Nobunaga, Nakamura and Imamichi (1966). This technique does not require Received 30 July 1986. Accepted 23 September 1988. anaesthesia and is relatively non-traumatic but it yields only small volumes and the vein can be difficult to locate in pigmented strains. Cardiac puncture has been widely used for collection of large blood samples and variations of the technique have been described by Cubitt and Barrett (1978), Frankenberg (1979) and Waynforth (1969). Performed under anaesthesia with the chest wall opened, cardiac puncture has a useful role in a terminal situation. Performed by needle puncture through the thorax under anaesthesia or in restrained conscious animals cardiac puncture enables large although variable volumes of blood to be taken. It can cause considerable trauma to the animal and deaths are not uncommon even in the hands of very experienced technicians. Repeated sampling endangers the animals to such an extent as to be unacceptable for long term toxicology studies. Blood samples taken by this method are often contaminated by tissue fluids. Most workers find the technique distasteful and acquiring skill could cause unacceptable suffering to the animals. A very commonly used vein for collection of blood is the external jugular. This vein can be cannulated under anaesthesia (Renaud, 1969), but it is possible to take samples by direct venipuncture in the depilated neck of rats. Very large volumes (up to 4 ml) can be taken and repeated sampling is possible. However, considerable skill is required and thrombotic complications can ensue. Perhaps the most widely used method for the collection of large blood samples involves puncture of the ophthalmic venous plexus or retro-orbital sinus (Stone 1954; Sanders 1963; Sorg & Buckner 1964). A siliconized microcapillary glass pipette
178 Conybeare, Leslie, Angles, Barrett, Luke & Gask is inserted under light anaesthesia. Large volumes can be taken but some trauma is involved and the procedure" cannot be repeated at frequent intervals. It requires consideral?le skill on the part of the operator who in the UK must hold an appropriate personal lincence under the Animals (Scientific Procedures) Act 1986. Training to acquire the skill can cause unacceptable suffering and many workers have ethical objections to the technique. The use of anaesthetics is undesirable for some blood biochemical and hormonal estimations. Tail tip amputation (the 'Farmer's Wife' technique) is easy to perform and requires minimal skill or training. Enta, Lockey and Reed (1968) and Stoltz and Bendall (1975) have recommended this method. Approximately 2 mm are cut off the tail tip, the animals are not endangered and do not appear to suffer, although tail necrosis can occur after frequent sampling. Relatively small volumes of blood can be taken (up to O' 5 ml from Wistar rats) and multiple sampling is limited by the length of the tail. Blood samples are inevitably contaminated by tissue fluids. Variants on this technique include cutting along the lateral caudal vein or artery. In our laboratories, for collection of blood samples in all studies except those involving the administration of compounds intravenously (when we use the lateral caudal vein for administration), we take blood from the lateral caudal vein. The major problem with the use of the laterai caudal vein is to ensure an adequate blood flow. One simple technique is to briefly warm the tail by immersion in a water bath (up to 45 C). This causes a local vasodilatation which permits the use of the caudal vein for intravenous administration but is not very useful for collection of large blood samples because the vasodilatatation is transient and is confined to the tail. Prolonged vasodilatation with increased caudal blood flow can be achieved by use of a heating lamp placed above the rodent cage. This warms the whole environment of the animals. A major disadvantage is the lack of temperature control and the danger of burns due to radiant heat. We have therefore improved the technique of collection from the lateral caudal vein by using a controlled warm environment to produce adequate tail blood flow and vasodilatation. Materials and methods Controlled warm environment By warming the total environment of the rat or mouse to temperatures around 40 C, a large blood flow through the tail is ensured since in these rodents the tail is a major thermoregulatory organ with a large surface available for heat loss. The resulting enhanced blood flow makes it possible to collect blood from the lateral caudal vein easily. The chamber used consists of a double skinned stainless steel washable and autoclavable box with internal dimensions 580mm high, 800 mm wide and 620 mm deep. The doors are made of a transparent acrylic material and fold completely back for easy access. Adjustable shelves allow the chamber to be used for cages in a range of sizes suitable for any species of small laboratory animal. It has the capacity to hold 30 Wistar rats in 6 standard cages of 5 animals or a smaller number of rats singly housed. Alternatively it can hold 12 standard mouse cages each housing 4 or 5 mice. The internal temperature can be maintained at ± I C in the range 33 to 45 C. A temperature of 40 C has been found suitable for most strains of rats or mice. The duration of exposure required for adequate increase in caudal blood flow is around IO to 15min but varies to some extent depending on the size of the animals. Warmed air enters the chamber through vents in the inner lateral walls and leaves through perforations in the ceiling. The electrical equipment and air warming system are in a detachable box mounted on the roof of the chamber and can easily be removed for sterilization with ethylene oxide. The chamber itself can be readily dismantled for cleaning and autoclaving. This makes it suitable for use in long term toxicity studies conducted in barrier facilities. The chamber is now manufactured commercially and is available from Harvard Instruments, Fircroft Way, Edenbridge, Kent, UK.
Blood collection from rodents 179 Catheter system and collection technique The use of a conventional hypodermic needle can lead to the formation of small clots due to turbulence in the 'dead space' at the base of the needle. This can be prevented by using heparinized needles but it is undesirable to use heparin for some purposes, e.g. haematology. A system with minimal 'dead space' and a wide bore to render the use of anticoagulants in the collecting catheter unnecessary is therefore used. The 21 gauge Abbott Butterfly set has been found most satisfactory (Abbott Laboratories, Queensborough, Kent, UK) with only 3 mm of the catheter tubing left attached to the needle. After collection of the required sample the Butterfly needle is withdrawn and the blood flow staunched by the application of a piece of tissue paper applied with brief pressure to the bleeding point. Mice are normally restrained in conventional 'tube' holders but for rats we have often found it most convenient to use manual restraint by a second technician who can also use his thumb as a tourniquet and can massage the tail to accelerate bleeding if necessary. We have used this technique during the last 5-6 years. Most of our studies have been in rats of the SK&F Wistar strain although the technique works just as well with larger rats such as the Sprague-Dawley. For mice we have sampled from the SK&F CD I albino strain and a pigmented mouse, the B6C3-F1 hybrid. We have taken blood from rats aged from 3 to 112 weeks and from mice aged 6 to 100 weeks. For routine haematology and clinical chemistry on rat studies samples of 2 ml are taken but samples in excess of 4 ml are possible. We have sampled weekly for up to 12 months and with smaller volumes daily for up to 30 days and for pharmacokinetic studies have taken up to 8 samples per hour. Such frequent repeated sampling has been limited to O 25 ml on each occasion. In mice we have taken weekly samples of 1ml and have also taken smaller volumes daily for 7 days. In no case have we lost an animal or caused serious damage to a tail in over 45 000 sampling occasions. Experimental comparison of haematological and clinical chemistry parameters In a comparison of a range of clinical pathology parameters 9 female SK&F Wistar rats weighing between 220 and 240 g had 2 ml blood samples collected by three techniques. On the first day after an overnight fast the lateral caudal vein was used. Seven days later a sample was collected under ether anaesthesia from the ophthalmic venous plexus. The anaesthesia depth was then increased and a further terminal sample was taken from the inferior vena cava. Samples of 1ml were taken from each rat on each occasion into ethylene diamine tetra-acetic acid for haematology determinations and 1 ml into heparin for clinical chemistry determinations. The following haematological parameters were measured using the Coulter-Counter (Model S Plus): red blood cell count (RBC); haemoglobin (Hb); haematocrit ratio (HCT); mean corpuscular volume (MCV); mean corpuscular haemoglobin (MCH); mean corpuscular haemoglobin concentration (MCHC); platelet count (PLT); white blood cell count (WBC). The following clinical chemistry parameters were measured using the Hitachi 705 Multichannel Analyser: alanine aminotransferase (ALT); aspartate aminotransferase (AS); alkaline phosphatase (AP); glutamate dehydrogenase (GLDH); total plasma protein (T. prot); albumin (ALB); creatinine (CRE); glucose (Glu). Prolactin assays Prolactin was assayed in rat blood samples collected by three techniques. Groups of 10 male SK&FWistar rats weighing between 250 and 280g were used. A blood sample of I ml was collected form the rats in each group and a serum sample was used for assay. Group 1 had the blood sample collected from the ophthalmic venous plexus under Halothane anaesthesia. In group 2 the sample was collected from the ophthalmic venous plexus under ether anaesthesia and in the third group the sample was collected from the lateral caudal vein of the rats without the use of anaesthetic. Prolactin (PRL) concentrations were measured
180 Conybeare, Leslie, Angles, Barrett, Luke & Gask in duplicate 50 Itl aliquot volumes of serum using a double-antibody radioimmunoassay (RIA) system. Specific reagents for rat PRL RIA were supplied by the Pituitary Hormones and Antisera Centre, Harbor-UCLA Medical Centre, Torrance, California, USA. The results are expressed in terms of RP-2 standards. The limit of sensitivity of this assay is 2 0 ng/ml with an inter and intra-assay coefficient of variation for the medium quality control of 7,4 and 12'5070, respectively. Results Haemat%gy Table 1 lists the mean parameters for the blood samples taken by the three collection methods. Blood collected from the ophthalmic venous plexus had slightly lower erythrocyte counts than Table 1. Mean haematology volumes (± SEM) Blood collection RBC technique x 10 12 /1 Hb (g/dl) HCT (Ratio) MCV (fl) MCH (pg) HCHC (G/dl) WBC (x 10 9 /1) Lateral caudal vein 8 15 ±0'20 16'0±0'3 0'427 ± 0,009 52'4±0'5 19'7±0'2 37'6±0'3 1020± 53 4'3±0'3 Ophthalmic venous plexus 7,66" ± O'l! 15'3 b ±0'1 0 405 ± 0,006 52'8±0'6 19'9±0'2 37'8±0'2 995'±66 4'7±0'5 Interior vena cava 8'09±0'l! 15'0 b ±0'2 0 456' ± 0,005 56 3'±0 6 18 5 d ±O'2 32'9 d ±0-2 1013± 73 4 1 ±0 2 n = 9 unless otherwise indicated. 'Significantly lower than for vena cava blood (P= <0'05). bsignificantly lower than for caudal vein blood (P= <0'05). 'Significantly higher than for blood from caudal vein or ophthalmic plexus (P= <0'001). dsignificantly lower than for blood from caudal vein or ophthalmic plexus (P= <0'001). 'n=8. All P values are derived from Student's t test. Table 2 Clinical chemistry values (± SEM) Blood ALT AST AP GLDH T. Prot ALB CK BUN CREAT G/u collection (lull) (lull) (lull) (lull) (gil) (gil) (lull) (mmol/l) (mmol/l) (mmol/i) Lateral 64,0 140-2 95'1 13'9 72,4 36 4 541' 8 92 71 9 9'51 b caudal vein ±6-7 ± 10'2 ±8'6 ±2'1 ±I'I ±0'3 ±66 ±0'35 ±2'9 ±0'16 Ophthalmic 57,8 98,6' 87,3 9,8 71 1 35 '2 d 300 8,33 67'3',8 8,73 venous plexus ±5'9 ±7'3 ±9'1 ± 1 7 ±0'8 ±0'8 ±62 ±0'33 ± 1-5 ±0-31 Inferior 53,4 90,0' 77,6 13,8 66'4 f 32 9 225 8 51 71-3 8 61 vena cava ±4'8 ±3'9 ±5 4 ±2'0 ± 1 3 ±0'5 ±20 ±O-l! ±I-I ±0'32 n = 9 unless otherwise indicated 'Higher than for other two samples (P= <0'05). bhigher than for other two samples (P= <0-05). 'Lower than for caudal vein blood (P= <0-01). dlower than for caudal vein blood and higher than for vena cava blood (P= <0 05 'Lower than for vena cava sample (P= <0'05). flower than for other two samples (P= <0'01). 8n=8 All P values derived by Student's t test. and P= <0'01).
Blood collection from rodents 181 120 100 80 E... 0).5.60 c: t5 (ll a ~ 40 20 o......... orbit orbit sampling Fig. 1. Mean serum prolactin tail site levels for male rats. blood collected from the vena cava. Blood collected from the caudal vein had a significantly higher haemoglobin level than blood collected by the other two routes. In the blood samples taken from the vena cava the haematocrit, MCV, MCH and MCHC values were significantly different from those taken from the caudal vein or from the ophthalmic venous plexus. These statistically significant differences have not been confirmed in some other comparative haematological evaluations which we have performed. Clinical chemistry These results are shown in Table 2. Slight differences between the different blood samples were seen in this study. Although some of these were statistically significant other comparative studies have shown these variations are not consistent. Prolactin assays The results of these comparisons are shown graphically in Fig. 1. Blood sampled by the tail bleed method shows significantly lower prolactin levels than in blood taken under anaesthesia from the ophthalmic venous plexus. Despite the fact that no anaesthetic was used for the tail bleed sampling this low prolactin level indicates that little stress was caused by the sampling technique. Conclusions The method which we have improved has enabled us to take frequent blood samples of adequate size from rats and mice for hormonal studies, drug metabolism and pharmacokinetic studies and for haematology and clinical chemistry in long term toxicology studies in rats or mice. It causes minimal stress to the animals as shown by the low prolactin values reported here. There is virtually no risk to the animals. The range of clinical pathology values we have found in the study and in many other comparative studies which we have performed are consistently within the normal published range of such values for the rat and the mouse. Acknowledgments We wish to thank numerous colleagues in Pathology and Toxicology at SK&F Research Ltd, Welwyn for their contributions to this work. In particular, Dr C. G. Brown, Miss K. Fowler, Miss J. Moyes, Miss L. Reilly and Miss J. Bone. References Cocchetto DM & Bjornsson TD (1983) Methods for vascular access and collection of body fluids from the laboratory rat. Journal of Pharmaceutical Sciences 72 465-492 Cubitt 10K & Barrett CP (1978) A comparison of serum calcium levels by two methods of cardiac puncture. Laboratory Animal Science 28, 347
182 Conybeare, Leslie, Angles, Barrett, Luke & Gask Enta T, Lockey SD and Reed CE (1968). A rapid safe technique for repeated blood collection from small laboratory animals: the Farmer's Wife method. Proceedings of the Society for Experimental Biology and Medicine 127, 136-137 Frankenberg L (1979) Cardiac puncture in the mouse through the anterior thoracic aperture. Laboratory Animals 13 311-312 Nobunaga T, Nakamura K & Imamichi T (1966) A method for intravenous injection and collection of blood from rats and mice without restraint and anaesthesia. Laboratory Animal Care 16, 40-49 Renaud S (1969) Jugular vein technique for blood colleciton and intravenous injection in the rat. Laboratory Animal Care 19, 664-665 Sanders BJ (1963) The application of the orbital sinus bleeding technic to germ free animals. American Journal of Clinical Pathology 40, 46-47 Sorg DA & Buckner B (1964) A simple method of obtaining venous blood from small laboratory animals. Proceedings of the Society for Experimental Biology and Medicine 115, 1131-1132 Stoltz DR & Bendall RD (1975) A simple technique for repeated collection of blood samples from mice. Laboratory Animal Science 25, 353-354 Stone SH (1954) Method for obtaining venous blood from the orbital sinus of the rat or mouse. Science, New York 119, 100 Waynforth HB (1969) Animal operative techniques (in the mouse, rat, guinea pig and rabbit). In Techniques in Protein Biosynthesis (eds PN Campbell and JR Sargent), vol. 2. London: Academic Press