The commonest biopsies to perform in practice are those of the skin, liver, muscle, intestine and bone marrow although other tissues such as kidney, spleen, lung or undefined masses might also be biopsied more rarely.

Skin
Skin biopsies are generally collected using specifically designed punches (e.g. Stiefel) that are no less than 6 mm diameter. Where nodular skin lesions are to be sampled that are covered by normal skin then standard surgical preparation can be performed. However, if there are any visible skin changes such as scaling, crusting or vesicles then it is imperative that no preparation or cleaning of the skin takes place other than perhaps a light alcohol spray. Even light cleansing of such lesions may remove superficial key diagnostic features preventing a histopathological diagnosis. Sedation is generally advisable followed by local anaesthetic infiltration under the skin lesions or as a ring block taking care not to disrupt the skin lesions. Where large or multiple lesions exist then it is preferable to collect several biopsies (e.g. 4 - 6) from different sites. The circular biopsy punches can be 'drilled' into the skin to the full depth of the dermis. For histopathological examination then placement of the biopsy in standard 10% neutral buffered formalin should follow immediately after collection. If larger excisional biopsies are collected then these should be pinned flat on cardboard prior to immersion in the formalin. If culture is required from the biopsy then placement in an empty sterile plain blood tube (red top) along with a drop of sterile saline should preserve bacteria/fungi en route to the laboratory.

Liver:
Transabdominal ultrasonography should be used to guide liver biopsy and a site is usually chosen based on thickness of imaged hepatic tissue, absence of large vessels and, occasionally, focal presence of hepatic tissue with an abnormal ultrasonographic appearance. Most biopsies will be collected from the right hemithorax between the 10th and 14th intercostal spaces.
The main theoretical adverse effects of liver biopsy in the horse are haemorrhage, colic, peritonitis, pleuritis and pneumothorax. In the author's experience all of these problems are very rare and the requirement for pre-biopsy coagulation assessment is questionable and never performed by this author.
Biopsy needles of 14 gauge and no less than 10 cm length, but preferably 16 - 20 cm are best. Generally 3 types of biopsy device are possible: a simple hand-operated device such as the 'Trucut' needle; a semi-automatic spring loaded biopsy needle device; or a separate biopsy gun device into which a simpler needle can be loaded.
Following sedation the biopsy site is prepared for a sterile procedure and 5 - 10 ml local anaesthetic is infiltrated subcutaneously and through the intercostal muscles. A small stab incision is made and the biopsy needle is advanced into the liver and the biopsy is collected. Needles operate best when perpendicular to the skin although aiming the needle slightly cranial and dorsal will maximise the hepatic target. Biopsy specimens are placed in 10% neutral buffered formalin for histopathological examination and/or plain sterile containers for bacteriological culture. Phenylbutazone and a prophylactic antimicrobial are administered.

Muscle:
Exertional myopathies are best investigated with type 2 rich semimembranosus biopsies, whereas nonexertional myopathies should be targeting the type 1 rich sacrocaudalis dorsalis. Following sedation and sterile preparation 10 - 20 ml local anaesthetic is injected subcutaneously. A 5 - 7 cm skin incision is made and subcutaneous tissue is cut to expose underlying muscle. Two parallel incisions are made into the muscle parallel with the muscle fibres and about 1 cm apart. This is undermined with curved artery forceps and clamped at each end before excising with scissors. Closure is routine. The muscle is best pinned onto card or a wooden tongue depressor before placing into 10% formalin. Depending on the examinations required, a section of muscle might be preserved without formalin for further analysis.

Intestine:
Various parts of the intestine may be affected by disease processes and the site of choice may well depend on the clinical presentation.
Rectal biopsy is the easiest and most accessible part of the gastrointestinal tract but clearly it is only likely to be pathologically affected in cases showing signs of distal intestinal tract disease (i.e. diarrhoea). It is easy to perform with light sedation. Biopsy forceps are the best and most appropriate tool with which to collect the rectal biopsy although beer bottle caps have been used by some practitioners with success by pinching mucosa between the cap and the thumb. The site to choose is in the dorsal midline. The biopsies may be put into formalin for histopathological analysis and/or cultured for enteropathogenic bacteria such as Salmonellae.
Duodenal pinch biopsies can be easily taken via duodenoscopy under sedation although they inevitably tend to be fairly superficial and damaged by crush artefact. Full thickness small intestinal biopsies require either laparoscopy in the sedated horse or laparotomy under a general anaesthetic.
Full thickness colon biopsies require laparotomy under general anaesthetic

Bone marrow:
Bone marrow biopsy is a relatively straightforward procedure in horses. The preferred site in adult horses is the midline sternum level with the points of the elbows in a horse standing square. In foals, the tuber coxa can be used. Following sedation and sterile preparation, 5 ml of local anaesthetic is infiltrated subcutaneously and in the tissues to the depth of the midline sternum. A small stab incision is made with a No. 15 scalpel blade. An 8 gauge Jamshidi needle is then inserted and drilled into the bone. This collects a small core of bone marrow which is preserved in 10% neutral buffered formalin for a better evaluation of cells within bone marrow. Very occasionally no sample is obtained or a dense white intersternebral sample is collected in which case, another site is chosen as the needle may well be sitting in an intersternebral space rather than in an a sternebra itself.