Fig. 13.1
An artery containing amyloid stained by Congo red. In ordinary, unpolarised illumination, amyloid appears red
Even with satisfactory staining by Congo red, there may be staining of normal structures such as elastic laminae in arteries, or there may be overstaining of the background, and more elaborate microscopy is usually used to confirm that positive staining identifies amyloid, or to exclude amyloid, which may be missed on ordinary microscopy. The special technique is polarisation microscopy. The findings of polarisation microscopy will be described before the relevant principles of optical physics, supported by experimental evidence, are used to explain the scientific basis of those findings (Section ‘Principles of Polarisation Microscopy and Explanation of the Optical Properties of Amyloid Stained by Congo Red’) [3, 10, 11].
Polarisation Microscopy of Amyloid Stained by Congo Red
Most microscopes can be fitted with a polarising filter between the light source and the condenser, called the polariser, and also another polarising filter between the objective lens and the eyepieces, called the analyser. These filters convert ordinary, unpolarised light into light which is only vibrating in one plane, and do not allow light to pass if it is polarised perpendicularly to the plane of the filter. When the planes of the polariser and analyser are perpendicular to each other, the filters are said to be crossed, and no linearly polarised light produced by the polariser can pass the analyser. The microscope field now appears its darkest. Often, when descriptions are given of the conditions needed to study Congo red-stained amyloid, the terms ‘in polarised light’ and ‘on polarising microscopy’ and so on are used, but these should be understood to mean examination of a section between polariser and analyser [11].
When amyloid stained by Congo red is examined between crossed polariser and analyser, some appears bright and coloured (Fig. 13.2). This is best looked for on a microscope using maximum light intensity and minimum condenser aperture. If the section can be rotated on the microscope stage, the bright and dark areas in amyloid deposits will be seen to change positions, depending on their relation to the plane of the polariser.
Fig. 13.2
The artery in Fig. 13.1 examined between accurately crossed polariser and analyser. The background is dark. Some parts of the amyloid appear bright and coloured, with combinations of anomalous colours that could appear blue/green and yellow/green, or green and yellow, or blue and yellow. The different colours are in areas roughly perpendicular to each other
The colour is almost always said to be green or apple green, often qualified with the words characteristic or typical [11]. A pure green colour may occasionally be seen on an ordinary microscope, but this is usually by chance, and to produce pure green may need either use of a microscope specifically designed for polarisation microscopy, or insertion into an ordinary microscope of a piece of equipment called a compensator (Fig. 13.3). The function of a compensator is explained later (Section ‘Compensation’).
Fig. 13.3
The artery in Fig. 13.1 examined between accurately crossed polariser and analyser, with insertion of an elliptical compensator into the light path. Adjustment of the amount of compensation has nullified effects of strain birefringence in the optical system and has given a relatively pure green anomalous colour, with different amounts of brightness
More commonly, the colour is not pure green but a combination of colours. These, including green, are called anomalous colours. The combinations may appear blue/green and yellow/green, or green and yellow, or blue and yellow (Fig. 13.2). The different colours are seen in areas roughly perpendicular to each other, appear to rotate as the section is rotated and exchange positions when the stage is rotated by 90°. The colours are not always uniformly distributed across a section. Often, changing the objective lens to one of a different magnification changes the colours.
When either the polariser or the analyser is rotated from the position that gives the darkest background, the background becomes lighter, more of the amyloid stained by Congo red appears coloured and other anomalous colours appear. These are orange and light blue/green (Fig. 13.4), or purplish red and greenish white (Fig. 13.5), or other combinations. If these colours are seen when the polariser and analyser are supposedly crossed, this means that the polarising filters are not accurately crossed. In contrast, blue/green and yellow/green, or green and yellow, or blue and yellow may be seen with perfectly accurate crossing of the filters.
Fig. 13.4
The artery in Fig. 13.2 examined with slight uncrossing of polariser and analyser. The background is lighter, and the anomalous colours are predominantly orange and light blue/green
The orange and other anomalous colours change when the section is rotated or when the polariser or analyser is rotated either further in the same direction or in the opposite direction. Amyloid deposits pass through a mixed red and colourless appearance, and then, when the planes of the polariser and analyser are parallel, amyloid deposits appear red and virtually indistinguishable from the appearance in ordinary illumination. These changes in colour are a useful way to confirm amyloid.
Colours other than green are hardly ever reported in papers on Congo red-stained amyloid, even though they can be easily seen in published illustrations. Some figures show the colours produced by uncrossing the polariser and analyser, but the legends rarely mention this. Only 59 (31 %) of 191 published figures showed a pure green colour, including those given the benefit of doubt in equivocal cases. The others showed green and yellow or blue and yellow in 62 (32 %), green and red or green and a colour other than red or yellow in 38 (20 %), and various colours without any green or blue, either combinations or single colours, in 32 (17 %). In 127 (66 %) of these 191 figures, there was a discrepancy between the description of colours and their illustration, mostly because green alone was mentioned in descriptions, but green and other colours could be seen in figures, although the other colours were ignored. In 30 (24 %) of these 127, green was mentioned in descriptions, but there was no green at all visible in figures [11].
No paper blamed an inability to illustrate pure green, or to show any green, on failure of accurate reproduction of colours. Accordingly, the illustrated colours can be considered representative of what was seen and accepted by authors, reviewers and editors as diagnostic of amyloid [11]. One implication is that most people using Congo red to stain amyloid assume that green is essential for the diagnosis and that they should only report green, even if they do not see either green on its own or any green at all. The mention of green is considered sufficient for the diagnosis. Another implication is that any explanation of the optical properties of Congo red-stained amyloid must account for all the colours, and not just green.
All these optical properties arise from the orientation of Congo red molecules on amyloid fibrils and are identical in specimens in which Congo red is orientated in other ways, such as experimentally in smears of solutions of the stain. Congo red is orientated on cellulose in plant cell walls and sometimes on aggregated immunoglobulin light chains in casts in myeloma kidney, but these materials should be easily distinguishable from amyloid by their appearance and position within a section. Other things that commonly stain with Congo red, such as elastic laminae and eosinophils, do not align the molecules sufficiently to give anomalous colours.
Principles of Polarisation Microscopy and Explanation of the Optical Properties of Amyloid Stained by Congo Red
Birefringence
Parts of Congo red-stained amyloid appear bright and coloured between crossed polariser and analyser (Fig. 13.2). The brightness is because the orientation of Congo red molecules on amyloid fibrils makes the Congo red molecules birefringent. This means that they have an asymmetrical arrangement of components that affect the velocity of light as it passes through. The refractive index is the velocity of light in air or a vacuum compared with the velocity in a material. A birefringent material has two extremes of refractive index, because light passing through one axis travels more slowly, giving a higher refractive index, than light passing through the axis perpendicular to this. These are called the slow axis and the fast axis. The birefringence is measured as the difference between the refractive indices of the two axes.
When linearly polarised light passes through a birefringent material with its axes at 45° to the plane of the light, the light becomes elliptically polarised (Fig. 13.6). This means that an observer looking at the light source and able to detect just one light wave would see a change from a wave vibrating in one linear plane to a wave tracing an elliptical path. The size of the ellipse depends on the retardance, which is the birefringence multiplied by the thickness of the material. Thicker layers of a material give more pronounced birefringent effects. This is how thicker sections allow easier detection of these effects, and how the effects may not be detectable in thin sections. Some elliptically polarised light is able to pass the analyser because it is no longer only vibrating perpendicularly to the plane of the analyser, and the material appears bright against the dark background.
Fig. 13.6
Diagram showing production of elliptically polarised light from linearly polarised light, such as in orientated Congo red. When a light wave imagined to be coming towards the observer passes through a polariser and then through a birefringent material with its fast and slow axes at 45° to the plane of the polariser, there is a change from linear to elliptical polarisation. This allows transmittance of some light by a crossed analyser. For the retardances of Congo red, the retardance determines the size of the ellipse in the plane of the analyser and the orientation of the fast and slow axes determines the direction of rotation of the elliptically polarised light. Reproduced from Bull R C Path. October 2008;144:263–6 with permission
Elliptical polarisation occurs because a wave of polarised light entering a birefringent material with its axes at 45° to the plane of the light can be considered to be split into two vectors perpendicular to each other, one in the fast axis and the other in the slow axis. These take a different time to travel through the material. On leaving the material, the vectors recombine into one wave, but the tip of one vector lags behind the tip of the other vector in time and is vibrating in a different plane in space. The distance travelled in air by the tip of the vector in the fast axis before the tip of the vector in the slow axis emerges from the material is equal to the retardance, which can be measured.
Because the vectors are separated in time and space before recombining, the tip of their resultant combined wave is no longer vibrating only in the plane of the polariser, but appears to trace an elliptical path as it approaches the observer. This can be imagined to follow the shape of a corkscrew flattened from side to side, with either a clockwise or an anticlockwise helix (Fig. 13.6). Within the range of retardances that have ever been reported in Congo red-stained amyloid, the dimension of the ellipse in the plane of the analyser is proportional to the retardance. Larger retardances give a larger component in the plane of the analyser and potentially more transmittance of light by the analyser. For any particular retardance, the orientation of the fast and slow axes relative to the plane of the polariser determines the direction of rotation of the tip of the combined vector. This means that if the birefringent object is rotated by 90°, the ellipse has an opposite direction of rotation. Similarly, the direction of rotation is opposite in a part of the birefringent object with the fast and slow axes orientated at 90° to those elsewhere [3, 10].
This production of elliptically polarised light is the mechanism that explains how some light can be transmitted by a crossed analyser, rather than other suggested mechanisms, which are insignificant in Congo red-stained amyloid. One erroneous suggestion was a change of the plane of linearly polarised light by optical rotation, also called optical activity, but this is too slight to have any detectable effect in sections, no matter how thick [3, 10].
Birefringent effects are most marked when the fast and slow axes of the material are orientated at 45° between the planes of the crossed polariser and analyser, and decline away from this position, if there is rotation of either the material or a polarising filter, for example. Birefringent effects disappear when the axes are parallel and perpendicular to the planes of polarisation (Fig. 13.7). This is because the light can no longer be split into vectors in both the fast axis and the slow axis, and explains how only some parts of randomly orientated Congo red-stained amyloid appear bright between crossed polariser and analyser (Fig. 13.2).
Fig. 13.7
Renal tubules stained by Congo red. Ordinary microscopy shows amyloid distributed uniformly in the circular basement membranes. When examined between accurately crossed polariser and analyser, as illustrated, an anomalous green colour with different amounts of brightness is seen, except at the four points on the circular objects that correspond with planes parallel and perpendicular to the planes of the polariser and analyser. Pure green is only seen in optically perfect conditions. Reproduced from Lab Invest. 2008;88:232–42 with permission
A birefringent material that has no colour on ordinary microscopy, such as unstained collagen, usually appears bright white under these conditions, because all wavelengths of light are elliptically polarised to virtually the same extent and are transmitted equally by the analyser. A coloured birefringent material, such as Congo red-stained amyloid, appears coloured under these conditions, although the colours seen are usually different from those seen in ordinary illumination, and are called anomalous colours. These colours appear because there is a difference in how wavelengths of light are affected through the spectrum. There are two fundamental processes that interact to affect wavelengths transmitted by an analyser, which are absorption and anomalous dispersion of the refractive index, although another factor, compensation, is often a contributor to the final outcome.
Absorption
Absorption means a reduction in intensity of light. A coloured material has at least one absorption peak in the visible spectrum, and removal of the absorbed wavelengths from white light, or reduction of their transmittance, gives the observed colour in ordinary illumination. Congo red has an absorption peak at a wavelength of about 500 nm, which is in the blue/green part of the spectrum (Fig. 13.8). Removal of blue/green or green from white light gives red (Fig. 13.1). Usually, absorption in a coloured material only occurs in one axis, and still occurs when that axis is at 45° to the plane of the polariser, although the amount of absorption is half that of the maximum absorption, when the axis is parallel to the polariser plane. When the absorbing axis is perpendicular to the polariser plane, the absorption is least. Variation of the amount of absorption depending on the plane of polarisation is called dichroism.
