Hepatic needle true-cut biopsies (especially 18-gauge) collected under ultrasonographic guidance may yield samples too small and fragmented for accurate diagnosis because of a lack of representative acinar units (at least 15 portal triads should be sampled). Furthermore, needle biopsies are usually only collected from the more safely sampled left-side lobes, which may miss lesions differentially affecting liver lobes (eg, cholangiohepatitis in cats). Blind-needle biopsies done without ultrasonographic guidance are hazardous and ill advised in animals with suspected hepatic hilar or mesenteric lymphadenomegaly; involvement of the common bile duct, gallbladder, intestines (eg, inflammatory bowel disease, infiltrative disease), or pancreas; or multiple organ abnormalities. An exploratory laparotomy is more appropriate. When possible, wedge biopsies or laparoscopic cup forcep biopsies are preferred, because samples of adequate size can be easily and safely acquired from multiple liver lobes, ensuring accurate disease representation. These methods also permit assessment of the gross appearance of the liver. Notably, laparoscopic methods are not recommended when disease of the common bile duct or gallbladder is suspected that may necessitate a decompressive biliary procedure, cholecystectomy, or cholestotomy. Liver biopsy should always be done even if an obvious biliary abnormality is the predominant disease process, because underlying histologic liver lesions may indicate another primary disease process. It is also important to biopsy grossly normal liver when focal lesions are identified. This practice ensures 1) characterization of "normal" liver histology, 2) determines whether an underlying liver disease coexists, 3) investigates histology of liver distant to the gallbladder in animals undergoing cholecystectomy, and 4) provides multiple liver lobe samples needed to confirm suspected MVD (because this lesion is variable among liver lobes)..
Routine biopsy evaluation should include examination of a cytologic imprint, Gram stain (if suppurative or pyogranulomatous inflammation is cytologically detected), routine H&E staining and interpretation, as well as staining with a reticulin stain (discloses sinusoidal collapse of the supporting scaffolding of the liver), Masson's trichrome (to confirm presence and severity of fibrillar connective tissue deposition), Prussian blue to identify iron retention in Kupffer cells (fixed macrophages) and hepatocytes (helps confirm lobular involvement in inflammation, Kupffer cell activation, presence of rare hemochromatosis), rhodanine stain for copper (confirms and can be used to quantify copper with digital scanning of the rhodanine-stained section), aerobic and anaerobic bacterial cultures of liver and bile, and quantification of liver metals (copper, iron, and zinc concentrations because these values can help evaluate risk of oxidative injury and need for zinc supplementation). A tissue sample (formalin fixed) should also be reserved for other special case-specific studies such as special immunohistochemical stains or for PCR testing for infectious agents.
Before biopsy, bleeding tendencies should be evaluated by careful review of the history, physical examination, blood smear (to confirm platelets ≥100,000/μL), routine coagulation profile (prothrombin time [PT], activated partial thromboplastin time [APTT]), von Willebrand factor (vWF) activity in high-risk breeds, and a buccal mucosal bleeding time. Routine coagulation assessments have low reliability to detect bleeding risk. The buccal mucosal bleeding time is more relevant when performed immediately before the procedure. Animals suspected of having bleeding tendencies should be treated with vitamin K1 (1–5 mg/kg, PO or SC, at 0, 12, 24, and 36 hr, and recheck coagulation at 2 days) before tissue sampling. If buccal mucosal bleeding time is >5 min, a fresh frozen plasma transfusion is indicated, as is administration of desmopressin acetate (DDAVP, 0.3–1 mcg/kg diluted in saline, slow IV, over 15–30 min), which increases plasma vWF 2-fold over baseline within 1 hr as well as plasma activity of Factor VIII. DDAVP can initiate a hemostatic effect in dogs with type 1 vWF (partial quantitative deficiency) but not in dogs with qualitative defects or complete vWF deficiency. In many people with liver disease, DDAVP has improved coagulation ability, although the exact mechanisms remain incompletely clarified.