Skip to main content

Our PhD Studentship awards are open for outline applications! Awards of three/four years are available including joint awards with the BHF.

NC3Rs | 20 Years: Pioneering Better Science

Blood sampling: General principles

Principles to help minimise stress and discomfort in laboratory animals when taking blood samples.

Approvals, training and competency


  • Blood collection procedures should be approved by the institutional ethics committee (AWERB, IACUC or equivalent) and detailed in the approved study protocol.
  • In the UK, blood sampling is considered a scientific procedure when the sample is taken for a scientific purpose. Removal of blood can itself have adverse consequences to the animal, the pain caused by the method of sampling (i.e. insertion of the needle into the skin) is deemed to exceed the level above which work needs to be regulated by the ASPA.
  • A project licence should detail the site, volume and frequency of blood sampling. These should be justified in terms of the science and animal welfare.
  • A personal licence is also required for the person taking the blood sample. Blood collection techniques can only be observed and not practised before applying for a personal licence. Once a personal licence is obtained, blood collection can only be carried out under supervision until the new licensee is deemed competent in the particular technique/s.

Training and competency

  • Competence in animal handling and restraint is an essential pre-requisite to blood sampling. Please see our dedicated resource on handling and restraint.     
  • Training should be done by appropriate, competent trainers so that the most refined and up to date methods are passed on. 
  • Criteria for competency should be defined for each technique. For example, competency in blood sampling from the canine cephalic vein would involve demonstrating:
    • Knowledge of the technique (e.g. suitability for purpose, appropriate needle size, permissible collection volumes, potential adverse effects and how to cope with them, how to handle the blood sample once taken).
    • The ability to restrain the dog calmly.
    • The ability to suitably prepare the sampling site - including hair removal and skin preparation in a way that does not distress the dog or damage the skin.
    • Accurate location and good dilation of the vessel.
    • Insertion of the needle without causing distress to the dog.
    • Removal of a non-haemolysed blood sample at an appropriate speed and volume without causing bruising.
  • Before attempting to perform any techniques, trainers should discuss with trainees the expectations regarding competency in the particular technique and how the skills will be assessed. Factors which will determine competency include assessment of protocol compliance, attitude and empathy and consideration for health and safety as well as competence in the practical skills required for blood sample collection. 
  • The amount of training and practice required to achieve a given level of competence in a particular technique varies from individual to individual depending on, for example, manual dexterity, prior experience, attitude, and the skills of the instructor.
  • To optimise sample quality and to minimise stress to the animals, blood sampling should preferably be carried out by an experienced person who does the procedure frequently and maintains their skills. Retraining or additional supervision is necessary if a technique is not conducted routinely and individuals should be encouraged to ask for help where necessary.
  • Inexperienced persons should first examine dead animals (euthanised for another purpose) to learn the relevant anatomy and thereby avoid having to make repeated unsuccessful entries when trying to locate a blood vessel. Use should be made of demonstration and instruction videos, such as those in this resource and on the Norecopa website. Inanimate objects (e.g. oranges) and imitation training aids/veterinary simulators (e.g. CurVetTM Rat Training SimulatorMimicky Mousecanine head and foreleg models) can be used to gain familiarity in handling and using needles and syringes, before carrying out any work on live animals. Observing experienced personnel will also help in learning the technique.
  • DOPS (Direct Observation of Procedural/ Practical Skills) are a well-defined mechanism to measure the competence of an individual in a procedure or practical task. The LASA DOPS website contains DOPs for blood sampling from a variety of mammalian species.

Practical considerations

  • Invest time in habituating animals to procedures, which is possible in a few short sessions. This will improve the experience of the animal and handler and will lead to obtaining a better-quality blood sample. Well-habituated animals may be able to be sampled unrestrained.
  • Prepare ahead of time with labelled collection tubes and any necessary equipment for processing samples (e.g. ice, tube roller/mixer, centrifuges etc). This is particularly important for procedures such as pharmacokinetic studies where multiple blood samples are taken in close succession.    
  • Arrange any required transport/courier services in advance.   
  • Plan ahead to ensure sufficient staff resource is available to assist with the sampling procedure. This is particularly important, for example, if manual restraint rather than a restraint device is to be used or if large groups of animals are to be sampled.



  • To prevent excess vacuuming on a syringe, consider the use of vacutainer blood collection tubes, which create a vacuum seal inside the tube to facilitate the draw of a pre-determined volume of blood. This reduces the likelihood of haematoma formation, particularly in larger animals such as dogs and primates. 
  • An advantage of using a vacutainer is that the blood does not need to be transferred from a syringe to a collection tube. In addition, vacutainers may contain additives to stabilize and preserve the blood sample.

Warming boxes

  • Blood sampling of rodents may require the animals to be warmed to dilate blood vessels prior to sampling. 
  • Best practice is to thermostatically control the temperature at no more than body heat. A timer should be used to ensure that the animal is warmed for no more than five minutes.
  • Animals in warming boxes should be should be kept under constant observation to prevent overheating. Signs of thermal stress include breathing more rapidly, panting, salivating, shaking the head or redness in the nose or muzzle. 
  • Warming boxes should not use a radiant source such as a heat lamp, and temperatures greater than body heat for a shorter period are not recommended due to the risk of hyperthermia. The warming box should be calibrated regularly.
  • Warming methods are contra-indicated for pregnant animals or in circumstances where there are significant adverse effects (e.g. where the animals have reached or are nearing the humane endpoint of the study).
  • Some laboratories apply local heat to the tail (e.g. using a heating pad or warm water bath) to avoid the need to heat the whole animal (which is likely to cause greater stress).  The temperature should still be no more than body heat.
  • Beware of the duration that small rodents are warmed, particularly when a group or cage of mice is placed in the warming chamber together but then sampled serially over a period of several minutes. For example, place a pair of mice at a time in the warming box so that no mouse spends more than a total of five minutes within the box.

Restraint devices (rodents)

  • Ideally animals should be habituated so that minimal or no restraint is necessary. In addition to improving the experience of the animal and handler, better quality blood samples are obtained when stressors are minimised.
  • Manual restraint of rodents is preferable to using a restraint tube, especially for animals that have been warmed.
  • Care should be taken when using a restraint device after warming, as the animal is more likely to experience syncope (collapse) or thermal stress when confined in the restrainer. 
  • Where a rodent restraint tube is used, it should be appropriate for the size of the animal in order to avoid damage to the tail, testes and limbs. It is important it is not too large or too small as either may result in injury.  
  • The restraint device should be correctly adjusted for the individual animal.
  • Non-rigid or cone shaped restraint devices may reduce the risks associated with restraint tubes.
  • Inexperienced operators should always be supervised when restraining a rodent using a device. 
  • All forms of restraining equipment should be frequently washed to prevent pheromonally-induced stress or cross-infection.

Aseptic technique

  • Hair around the sample collection site should be clipped and the skin aseptically prepared with a water-based chlorhexidine solution (or similar). Shaving with a scalpel blade is not recommended as it removes the epidermal layers of the skin. 
  • Non-toxic hair removal cream (Nair®, Veet®, etc.) may be considered for areas where clippers are difficult to use. Applications of such agents should be made in accordance with the manufacturer’s directives by placing a layer upon the area to be depilated for the designated time-frame. Care should be taken to ensure that the animal does not lick the cream off. The depilatory and hair is removed by wiping the area with a water-moistened gauze pad or cloth. It is important that all traces of the depilatory cream be removed in order to avoid possible irritation from excessive exposure to the active chemical agent. 
  • Needles should always be single-use to avoid them becoming contaminated (see our pages on needle reuse). Blunt and used needles should be disposed of directly into sharps containers.  


Handling and restraint

  • Please see our dedicated resource on handling and restraint.
  • Firm, empathetic handling is very important, as is the time required to withdraw the blood sample. Both these parameters can affect the degree of stress for the animal and consequently the quality of the sample and research data.
  • The animal should be restrained by an experienced person (preferably one known to the animal, especially for larger laboratory animal species). The correct level of restraint is that which allows a satisfactory sample to be taken at the first attempt but which does not cause the animal to become unnecessarily distressed.
  • Inanimate restrainers can be used, although these may not always be the best method for individual animals. Manual restraint facilitates recognition of distress more effectively and allows modification of the restraint in response to the animal’s requirements. Furthermore, restraint may not be needed if refined techniques such as, for example, positive reinforcement training or microsampling are used.
  • A vein will collapse if a sample is taken too quickly, so care should be taken to ensure that blood is withdrawn at an appropriate speed.
  • Depending on the species, consideration should be given to offering a reward after each bleed.
A mouse that has been acclimatised to be handled by a technician and has been trained to undergo tail vein blood sampling, can be handled without the need for restraint and without any obvious stress or discomfort.

Needle size

  • A sterile needle (or blood lancet) should be used to puncture the skin and underlying blood vessel.
  • The size of the needle (length and bore) is very important.
  • It is recommended to use a needle with sufficient diameter to ensure rapid blood withdrawal without collapsing the vein, within the constraint of avoiding haematoma (i.e. the bore should be just less than the diameter of the vessel).
  • As superficial blood vessels lie close to the skin surface, needles used to sample from these veins should be kept to a minimum length. This will ensure the best possible control of the needle during insertion, reducing the risk of injury to the animal.
  • Recommendations on appropriate needle size are given for specific techniques featured within these pages. A maximum gauge of 25G is recommended for rodents.

Site and location of the vein

  • Draw your sample from a visible vein. Sampling from a visible, superficial vein should reduce trauma and ensure a high quality blood sample is taken. 
  • Note that samples taken from different sites may have differences in biochemical / haematological values and so the same vessel should be used if samples are collected throughout a study (e.g. always from a tail vein). If repeated sampling is necessary, it is important to choose a vessel that will allow you to alternate between sample sites to avoid bruising etc.
  • Time should be spent accurately locating and, if necessary, manually dilating the vein through the use of pressure before puncturing the vessel.
  • It is recommended that no more than three attempts should be made to collect a blood sample by the venepuncture route. If attempts to collect blood are unsuccessful, the animal should be left to recover and sampling attempted again at a later time and preferably by a different individual.

Dilation of the vein

  • In conscious rodents, blood can be more easily obtained if the animal (or part of the animal the sample is taken from, e.g. tail) is warmed first following best practice (above).  
  • The use of topical local anaesthetic (e.g. EMLA, LMX) provides effective analgesia to prevent discomfort and pain on venepuncture. However, 15-30 minutes contact time is required for these compounds to have full effect. To prevent animals licking the cream off once it is applied, the area can be covered with an occlusive dressing (though in small rodents, this may present practical difficulties). 
  • Anaesthesia is not normally necessary for venous access, since the associated stress would probably be greater than the discomfort of a needle prick or of a puncture with a blood lancet.
  • Consideration should be given to bleeding when animals are anaesthetised for another purpose, however veterinary advice should be taken on how anaesthetics may affect blood sampling.
  • Some anaesthetics or components of anaesthetic mixtures (e.g. medetomidine or xylazine) cause vasoconstriction and so should be avoided. 
  • Some anaesthetics contain vasodilators, negating the need to warm the animal for vein dilation. Extra care should be taken to ensure bleeding has stopped once the sample is obtained.

Potential adverse effects

Blood sampling should not cause adverse effects at the site of the sampling. Potential problems due to poor sampling technique include stress, haemorrhage, bruising, thrombosis, infection at the site of needle entry, phlebitis, scarring and nerve damage.  It is a common misconception that these adverse effects are normal. Advice on treatment for adverse effects should be sought from the veterinary team and additional training provided.

  • Haemorrhage due to poor haemostasis is not a common problem unless the animal has a clotting defect, and in some cases, gentle continuous pressure applied for several minutes is all that is needed to stop the bleeding. Longer compression of the puncture site may be needed to stop bleeding following arterial sampling.
  • 'Bruising' is due to subcutaneous bleeding at the time of venepuncture or after the animal has been placed in its cage or pen, when the site might be aggravated by the animal itself through licking or rubbing. The animal should be checked after approximately 30 minutes and, if necessary, appropriate action taken (e.g. consult the Named Veterinary Surgeon).
  • Thrombosis (clotting) and phlebitis (inflammation of the vein) are usually caused by failure to employ aseptic technique. Occasionally they can result in self-mutilation.

Arterial puncture

The main reason for collecting blood from arteries is that large samples can be obtained rapidly and relatively easily. Many of the principles described above for venepuncture also apply to arterial puncture.


Catheterisation (also known as cannulation) is an important technique for removal of blood because it reduces the stress of multiple sampling associated with, for example, repeated restraint and needle sticks.

  • Catheterisation should be considered when repeated samples are required, especially over relatively short time periods.
  • In some species, it may be necessary to restrain the animal in some way to stop it removing the catheter. For example, rats are often restrained by a harness, swivel and tether system, which restricts normal movement. Animals should be acclimatised to any restraint system before catheterisation.
  • Tethered animals are often housed singly, thus adding to the stress and severity of the procedure. When dealing with social animals, every effort should be made to keep them in social groups. Catheterised pigs, cats and marmosets can be group-housed successfully with appropriate bandaging and protection for the catheter.
  • Catheterisation has the potential to cause discomfort to the animal and therefore warrants post-operative administration of analgesics and careful post-operative care and monitoring for the duration of time the cannula is in place.
  • Catheter-associated infections can be avoided through the use of sterile equipment and solutions, and by employing aseptic technique.
  • When selecting an animal for catheterisation they should be assessed on health status as well as behavioural temperament. This will help ensure that the animal can cope with the additional handling, catheter maintenance and dosing procedures required. 

Please see our dedicated resource on vascular catheters.   

Cardiac puncture

  • Cardiac puncture should only be carried out under deep terminal anaesthesia or on euthanised animals.
  • It is common to exsanguinate animals under terminal anaesthesia using this method. Where the procedure is intended as terminal, death after exsanguination should be ensured via an appropriate killing method as detailed in Schedule 1 of the ASPA.

Volume of blood to be removed

The volume of blood removed and the frequency of sampling should be based on the purpose of the scientific procedure and the total blood volume of the animal. It is essential to take account of the combined effect of sample volume and the frequency of sampling. If too much blood is withdrawn too rapidly, or too frequently without replacement, an animal can go into short-term hypovolaemic shock and/or in the longer term suffer anaemia (see below). Data interpretation and scientific validity may be confounded if excessive sampling is employed.

  • Sample volumes and number of samples should be kept to the minimum necessary. 
  • As a general guide, up to 10% of the total blood volume (TBV) can be taken on a single occasion from a normal, healthy animal on an adequate plane of nutrition with minimal adverse effects; this volume may be repeated after three to four weeks. Where blood is sampled repeatedly (e.g. weekly), the suggested limit is typically 15% TBV over 28 days. For repeat bleeds at shorter intervals (e.g. over a few days), a maximum of 1.0% TBV every 24 hours is recommended (also see our resource on microsampling). The effects of stress, site chosen and anaesthetic used, should be taken into account.
  • If frequent samples are necessary, the use of catheterisation as a less stressful alternative to repeated venepuncture should be considered. 

As a general rule, total blood volume can generally be estimated as 55 - 70 ml/kg body weight. However, care should be taken in these calculations as the percentage of total blood will be lower (-15%) in obese and older animals. Note that blood volume limits indicate the upper limit for a normal, healthy animal and that sample volumes should always be kept to the minimum necessary.

Below is a comprehensive table of total blood volumes and blood sample volume limits that are considered generally safe for healthy laboratory animals, domestic species and non-human primates (adapted from Wolfensohn S and Lloyd M 2003 [1]):

Species Reference weight (g) Blood volume (ml/kg)$ Total blood volume (TBV), normal adult (ml) Safe volume for single bleed (ml)* Bleed out volume (ml)    
Mouse 18 - 40 58.5

1.5 - 2.4

1.0 - 2.4

0.1 - 0.2

0.8 - 1.4

0.6 - 1.4

Rat 250 - 500 54 - 70

29 - 33

16 - 19

2.9 - 3.3

1.6 - 1.9

13 - 15

7.5 - 9

Hamster 85-150 78

6.3 - 9.7

7.1 - 11.2

0.6 - 0.9

0.7 - 1.1

2.9 - 4.5

3.3 - 5.2

Gerbil 55 - 100 66 - 78

4.5 - 7

3.8 - 6

0.4 - 0.7

0.4 - 0.6

2.2 - 3.5

1.9 - 2.9

Guinea pig 700 - 1200 69 - 75

59 - 84

48 - 63

6 - 8

5 - 6

29 - 42

24 - 31

Rabbit 1000 - 6000 57 - 65 58.5 - 585 5 - 50 31 - 310
Ferret 600- 2000 70 42 - 140 4 - 14 21 - 70
Dog - 70 - 110# 900 - 1170a 90 - 110 -
Cat - 47 - 65 140 - 200 14 - 20 -
Pig - Large white - 56 - 69 13,200 - 15,000 1320 - 1500 -
Pig - Yucatan - 56-69 4200 - 4800 420 - 480 -
Sheep - 58 - 64 4060 - 4480 400 - 450 -
Goat - 57 - 90 3990 - 6300 400 - 630 -
Cattle - 60 27,000 - 36,000b 2700 - 3600 -
Horse - 75 33,750 - 45,000b 3375 - 4500 -
Marmoset - 60 - 70 21 - 24.5 2.1 - 2.4 -
Rhesus macaque - 55 - 80

420 - 770

280 - 630

42 - 77

28 - 63

Long-tailed macaque - 50 - 96

280 - 560

140 - 420

28 - 56

14 - 42


$ A blood volume estimate for a single species may not reflect differences among individual breeds or variations due to age, size, or illness
* Single bleed of 10% total blood volume
# Much breed variation
a Beagle
b Assumes adult weight 450-600 kg

The EFPIA/ECVAM good practice guide to the administration of substances and removal of blood also contains recommended mean total blood volumes and maximum blood sample volumes for species of a given bodyweight.

Signs of shock and anaemia

It is essential to be able to recognise the clinical signs of shock and anaemia and be able to take appropriate action.

  • Signs of hypovolaemic shock include a fast and thready pulse, pale dry mucous membranes, cold skin and extremities, restlessness, hyperventilation, and a sub-normal body temperature. The Named Veterinary Surgeon should be consulted immediately if shock occurs. If more than 10% of the total blood volume has been removed, a routine replacement with the same volume of warm (30-39oC) normal buffered saline constitutes good animal care.
  • Signs of anaemia include pale mucous membranes of the conjunctiva or inside the mouth, pale tongue, gums, ears or footpads (non-pigmented animals), intolerance of exercise and, at the more extreme level, an increased respiratory rate when at rest. Where there is concern about the development of anaemia, packed cell volume, haemoglobin level, red blood cell and reticulocyte counts should be monitored throughout the series of bleeds using the results from the first sample from each animal as the baseline for the animal.

Resources and references

  1. Wolfensohn S and Lloyd M (2003). Handbook of Laboratory Animal Management and Welfare, 3rd edition, Blackwell Publishing Ltd.
  2. Hoggatt J et al. (2016). Bleeding the laboratory mouse: Not all methods are equal. Experimental hematology 44(2): 132-37. doi: 10.1016/j.exphem.2015.10.008
  3. Diehl KH et al. (2001). A good practice guide to the administration of substances and removal of blood, including routes and volumes. Journal of Applied Toxicology 21(1): 15-23. doi: 10.1002/jat.727
  4. Morton DB et al. (1993). Removal of blood from laboratory mammals and birds. Laboratory Animals 27(1): 1-22. doi: 10.1258/002367793781082412
  5. McGuill MW and Rowan AN (1989). Biological effects of blood loss: implications for sampling volumes and techniques. Institute of Laboratory Animal Research Journal 31(4): 5-20. doi: 10.1093/ILAR.31.4.5