To the Editor:

Transmission electron microscopy (TEM) has contributed considerably to liver research[1-10] after the introduction of perfusion fixation, with glutaraldehyde fixative through the portal vein. A clear distinction of sinusoidal endothelial, Kupffer, pit, and stellate cells became apparent, including details such as the endothelial fenestrae.[7]

Liver needle biopsies are taken to enable or support clinical diagnosis. To this end, paraffin sections are studied by light microscopy (LM) and sometimes a histochemical or immunohistochemical staining is added. TEM is not routinely used in diagnosing or investigating human liver disease. This might be the result of the use of immersion fixation for needle biopsies in both LM and TEM.

During immersion fixation the fixative penetrates the tissue by diffusion, and on its way the fixing compound is reacting with tissue components and is progressively exhausted. This leaves the center of the tissue unfixed for almost 1 hour[2] and only allows studying the well-fixed peripheral parenchymal cells. Sinusoidal cells and sinusoids collapse and lose details due to this improper fixation. As a result, it is impossible to study sinusoids and sinusoidal cells in human liver needle biopsies. This is an important disadvantage, because these cells play a key role in several pathophysiological mechanisms, such as cold ischemia,[11] drug-induced toxicity,[12] lipoprotein clearance,[13] and fibrosis.[14, 15]

There is a pressing need for properly fixed TEM material, as until now sinusoidal cells are mainly studied in human cell culture or animal models. In order to avoid the shortcomings of immersion fixation, we developed a new fixation technique for human liver needle biopsies equaling the quality of perfusion fixation.

Percutaneous liver biopsies with a 16G needle provide tissue specimens of 1-2 cm in length, with a diameter of 1-1.8 mm. This biopsy is immediately transferred to physiological saline with heparin at 37°C in a Petri dish, where it is wrapped in gauze over the full length of the biopsy. The gauze is gently closed on one side by an artery clamp (Fig. 1A,B). Compression on the tissue is avoided, since this leads to compression of the sinusoids and inhibition of perfusion. This approach also makes it easy to handle the biopsy, which is at the same time accessible to fixative and other fluids.


Figure 1. Schematic representation of jet-fixation. (A) Jet-fixation tripod with gravity-mediated perfusion from a bottle at 80 cm height with glutaraldehyde fixative at 37°C. (B) In detail, the liver biopsy is surrounded by a gauze (glass curtain), which is wrapped around the length of the biopsy and which is gently closed on one side by an artery clamp.

Download figure to PowerPoint

A bottle with glutaraldehyde fixative, 1.5% in cacodylate buffer 0.067 mol/L, is mounted at a height of 80 cm (32 inches). The fixative flows by gravity through a straight tube without any obstruction, ending with an 18G needle (Fig. 1A). A jet stream of fixative is sprayed from a rectangular position onto the biopsy, the needle being gently moved backwards and forwards over its entire length. Spraying time is 2 minutes with a flow of about 100 mL/min. Halfway through the procedure, the biopsy is turned to spray the other side as well. This technique sprays the fixative with a constant pressure onto the liver tissue.

After spraying, the biopsy is immersed in glutaraldehyde fixative to allow the fixative to react with the tissue for a total of 20 minutes. After this, the tissue will be further processed in washing buffer, followed by buffered 1% osmium tetroxide fixative. Ethanol dehydration follows[2] and ends with embedding in Epon for LM and TEM. Critical point drying or HMDS (hexamethyldisilazane) drying can be applied after dehydration for scanning electron microscopy (SEM).[1]

Jet-fixation leads to well-fixed tissue with open sinusoids, free of blood cells (Fig. 2B,C). In contrast to immersion fixation, the fine structure, shape, and spatial relationships of parenchymal cells, sinusoids, and sinusoidal cells, such as endothelial, Kupffer, stellate, and visiting cells is well preserved (Fig. 2C) and can be studied in detail by LM, TEM, and SEM. Sinusoidal endothelial cells spread out into the endothelial lining consisting of cellular processes with interspersed sieve plates containing fenestrae (Fig. 2D). The space of Disse is clearly defined as an extravascular space containing parenchymal cell microvilli, various elements of the extracellular matrix, processes of stellate cells, and sometimes fragments of a thin layer of basal lamina underneath the endothelial lining. In these preparations the parenchymal cells could be studied throughout the entire biopsy (Fig. 2B).


Figure 2. Light micrograph of a liver needle biopsy, immersion fixation (A) versus jet-fixation (B), and the corresponding TEM data of jet-fixed liver tissue (C,D). (A) In immersion fixation, all sinusoids are collapsed and lose detail, including the endothelial cells. Original magnification 10× objective. (B) After jet-fixation, the sinusoids are open and there is perfusion fixation quality up to and including the core of the biopsy. Original magnification 10× objective. (C) Low-magnification TEM picture (800×) of a human liver needle biopsy after jet fixation. Note that all sinusoids are open (*) and show many different details of the sinusoidal cells, such as endothelial, Kupffer, and stellate cells. (D) High-magnification TEM image showing detail of the endothelial lining containing fenestrae (<). The space of Disse shows the presence of microvilli and different components of the extracellular matrix. In the endothelial cytoplasm microtubuli and small electron-dense granules can be recognized. Original magnification 25,000×.

Download figure to PowerPoint

The proposed novel technique of jet-fixation makes it possible to obtain high-quality perfusion-type of fixation of human liver needle biopsies. Apparently, the sponge-like structure of the tissue, resulting from the anastomosing network of sinusoids, together with the application of a jet of fixative on the outside of the tissue, allows the fixative to enter the tissue in a way comparable to perfusion through the portal system. Based on our series of different experimental fixation times, including fixative concentration and fixation buffer conditions, we conclude that the sinusoids are instantaneously flushed by the fixative, removing blood cells and plasma and reaching the center of the tissue rapidly at full strength. Ideally, these results should be compared with the gold perfusion standard, namely, perfusion through the portal vein, but this is not applicable to human liver. However, the present results are comparable to the results of perfusion quality obtained with wedge biopsies injected with glutaraldehyde.[2]

In summary, jet-fixation is an easily applicable and low-cost perfusion technique for human and other liver needle biopsies. It enables anyone to improve the study of fine structural details, the shape and the spatial relationship of cells in healthy and diseased human, or experimental animal liver by LM, TEM, and SEM at a wide range of magnifications (2-100.000× and more).

  • Celien Vreuls, M.D.1

  • Eddie Wisse, Ph.D.2-4

  • Hans Duimel, B.Sc.2

  • Kris Stevens, M.D., Ph.D.2

  • Fons Verheyen, Ph.D.2

  • Filip Braet, Ph.D.3

  • Ann Driessen, M.D., Ph.D.1

  • Ger Koek, M.D., Ph.D.4

  • 1Department of Pathology

  • Maastricht University Medical Centre

  • Maastricht, The Netherlands

  • 2Electron Microscopy Unit (CRISP) and Department of Molecular Cell Biology

  • Maastricht University Medical Centre

  • Maastricht, The Netherlands

  • 3School of Medical Sciences (Discipline of Anatomy and Histology)

  • Bosch Institute and Australian Centre for Microscopy & Microanalysis

  • University of SydneySydney, Australia

  • 4Department of Internal Medicine

  • Maastricht University Medical Centre

  • Maastricht, The Netherlands


  1. Top of page
  2. References