Fig. 14.1
Diagnosis of amyloid in fixed tissues after CR staining, as viewed under three different types of illumination (“triple illumination”), in order to demonstrate the detection of amyloid and the pitfalls that may be encountered therein. Five different tissues A–E were examined, and the same frame of each of the five was photographed under three different types of illumination: (a)–(c). A Diagnosis of renal amyloid in a normal renal tissue section (autopsy tissue). B Diagnosis of amyloid in a fatty tissue smear; the tissue is too thick. C Diagnosis of amyloid in a very thin section. D Fatty tissue smear with false-positive CRF. E Fatty tissue smear with a false-positive CR-stained section and a false-positive CRF. (a) Viewed in bright light for recognition of CR staining. (b) Viewed under CR fluorescent illumination (CRF). (c) Viewed under polarized light to show the presence of green birefringence (GB). The main point of this illustration is to show the diagnostic value of CRF in the diagnosis of amyloid with a high level of sensitivity and specificity and also for the exclusion of artifacts
In A (a), an autopsy tissue section of normal thickness (approximately 6 μm) displaying renal vascular amyloid is marked in red. In (b), the same frame is shown in CRF. Here, the entire amyloid-containing area shows bright orange-red fluorescence. In (c), the same frame is seen again under polarized light—here, the amyloid-containing areas only partially show the pathognomonic green birefringence (GB). Other sections of the amyloid-containing area are dark, show no birefringence, and are therefore said to be in the “polarization shadow,” as indicated by the arrows in A(c) [7]. While in (a) and (c), this particular area, containing amyloid deposits, shows clear evidence of staining/fluorescence, it is partially obscured in (c). However, by rotating the slide table further, this area can also be shown to display GB [7].
In B (a), the rim of the subcutaneous fatty tissue aspirate (FTA) shows a somewhat reddish area suspicious for amyloid, which in (b) is clearly seen to produce very bright orange-red fluorescence by CRF. The diagnosis of amyloid is verified in (c) through GB, which is seen only in some parts of the amyloid deposits, while other parts are in the polarization shadow. Thus, in Fig. 14.1b, again only CRF shows the full extent of the small amyloid deposits. Under these conditions, smaller amyloid deposits could have been missed without CRF. CRF is particularly useful in specimens that contain very bright whitish collagen (such as FTA), which may mask the GB in polarized light (see Figs. 11.1–11.4 in [7]). When GB is not visible under such conditions (i.e., when there is abundant bright collagen), amyloid cannot be reliably excluded. In contrast, since CRF can shine through moderately thick FTAs, a negative CRF result essentially excludes the presence of amyloid. Thus, CRF can add precision to the diagnosis of amyloid in tissue smears obtained from FTA or elsewhere.
C (a) shows renal biopsy tissue (of approximately 2 μm thickness) with very pale reddish amyloid that is hardly visible at all in bright light. Its full extent is only clearly visible through the application of CRF in (b) and is confirmed as amyloid by GB in (c). Here, again, amyloid deposits are only partially visualized by GB. Without CRF screening, smaller and, in particular, minute amyloid deposits can easily be overlooked.
D and E present FTAs that are prone to artifacts as a consequence of their thickness (such artifacts can also occur in tissue sections, but less frequently). Figure 14.1a–c shows that when CRF is negative, amyloid is, most likely, not present. However, when CRF is visible, it cannot be assumed that amyloid is present, unless its presence can be confirmed by GB. Thus, CRF must be verified by GB before the presence of amyloid is diagnosed. However, false-positive GB can also occur. In such instances, the concomitant absence of orange-red fluorescence by CRF (not shown in this report) is helpful. This underscores the central role of CRF for the precise diagnosis of amyloid in tissue sections.
In D (b), a bright, unidentified fiber displays an intense CRF signal but is not co-stained with CR in (a) and does not show GB in (c). Thus, the CRF result represents a false-positive signal.
Another artifact is seen in E (a), which shows a bright red deposit that, at the first glance, appears to resemble amyloid. However, since there is neither a CRF signal in (b) nor a GB signal in (c), the false positivity of the CR-stained section seen in bright light is apparent. In the left upper corner in E (b), there is a CRF signal, as indicated by a white arrow, which is false-positive since it yields neither a CR signal in bright light (a) nor a GB signal in polarized light (c). Summarizing the results obtained from Fig. 14.1, it can be concluded that CRF is very useful as a screening method for the detection of amyloid deposits derived from all of the chemically different amyloids including artificial amyloids produced in vitro [5, 7]. An area that is identified as positive by CRF needs to be subsequently examined for the presence of amyloid using GB since the increased sensitivity of CRF also increases its lack of specificity (John H. Cooper, cited in [12]). The lack of specificity, however, is not a problem, since the artifacts can be identified as such and immediately excluded by triple illumination of the same frame in a tissue section. This is easily achieved by changing the light source while the same tissue frame remains in place. By this means, amyloid can easily be detected with high levels of both sensitivity and precision [5, 7].
Other problems can be encountered in sections that have been submitted for a second opinion or an “expert” opinion. Thus, sections may be very thin (as used in nephropathology) or (rarely) too thick or slashed due to an inappropriate biopsy technique. Sometimes, submitted, prestained tissue sections can appear overstained with Congo red, when the whole section will polarize green without discrimination. This situation arises, most probably, from a missed or inadequate differentiation step or the use of an inappropriate staining solution. More severe still is overstaining with hemalum on thicker sections: the hemalum can conceal the CR staining and GB. In these cases, CRF can be helpful since, to a certain extent, it can also “shine through” the overstained areas as a result of its very bright fluorescence (see above [7]).
It should be kept in mind that the pathognomonic GB seen by polarization microscopy (with appropriate equipment) can be demonstrated only in sections that are cut within standard thicknesses. When sections are below 1 μm in thickness, the anisotropy turns to bluish white, and when the thickness increases beyond the standard thickness, the green turns to yellow green and, with further increase, to yellow orange and finally red. All these colors are “specific for amyloid” [12]. Therefore, the pathognomonic characteristic of amyloid is most properly “colored anisotropy,” while green birefringence is only a consequence of the standard thickness of paraffin sections as published by J. H. Cooper, reviewed in [7].
Sampling Error as a Major Pitfall
Sampling error is one of the most common pitfalls encountered in the diagnosis of amyloid and deserves a separate section since it remains largely unmentioned and underdiscussed. It is characterized by the perceived absence of amyloid in a biopsy tissue section in patients with amyloidosis . It occurs most often in small biopsies and, in particular, in the very early phases of amyloidosis, when amyloid is still sparse as a consequence of its uneven distribution in tissues and organs. Another important peculiarity that must be considered in the search for early amyloid deposits is that amyloid deposition may not begin at the same time in all organs or may not accumulate with equal speed and strength in all organs. This conclusion is based on the author’s personal experience of experimental amyloidosis in mice. Since there is an organ preference in certain amyloid types, when abdominal fat (or other, “safer” sites used for screening) is negative for amyloid, one should consider obtaining biopsies from the most affected organs, such as heart, kidney, liver, etc. [13].
In the early stages of amyloidosis, the main question in small tissue sections that contain few amyloid deposits is as follows: how representative is a single tissue section? This situation is presented schematically in Fig. 14.2 where a small biopsy has been sectioned and the individual sections examined for amyloid. In Fig. 14.2, sections “Introduction and Overview,” “Tissues and Congo Red Staining for the Diagnosis of Amyloid,” “Minute Amounts of Missed Amyloids,” “Identified and Classified,” and “Take-Home Lessons” would yield a false-negative diagnosis. Also, in the absence of the increased sensitivity provided by CRF, the sparse amyloid deposits seen in sections “Evaluation of Congo Red-Stained Sections and Pitfalls” could be easily missed, possibly even during expert evaluation. It is also clear from Fig. 14.2 that the statement “no amyloid” without further qualification is incomplete. Since a negative amyloid diagnosis from a single tissue section can never be conclusive, in cases where a single tissue section is found to be negative, ten or more additional sections from the biopsy should be examined [7, 13]. These examples should serve to raise awareness of the importance of sampling errors in general; sampling errors due to organ preference have also been reported [13]. Thus, a comment should be added that biopsies which contain a limited amount of amyloid may be subject to sampling error (as illustrated in Fig. 14.2).
Fig. 14.2
Sampling error. A schematic illustration of how sampling error can occur in small biopsies with an uneven distribution of amyloid (seen most commonly in early amyloidosis). In this example, all sections of a small biopsy have been examined for amyloid (in red color) and presented in an enlarged view
Sampling error should also be considered when measuring the sensitivity of different methods for diagnosis of amyloid in tissue sections, and to avoid sampling errors, the same sections should always be evaluated when conducting interlaboratory comparisons (see Figs. 14.3, 14.4, 14.5, and 14.6). Finally, tissue sections that are prone to sampling errors, and which contain areas with variable amounts of amyloid, offer the possibility of measuring the sensitivity of various amyloid detection methods as shown in Fig. 14.5c (discussed below).
Fig. 14.3
Quality of diagnosis of amyloid (shaded area) shown by comparison of the results from three different laboratories using retrieved and reexamined biopsies of children with early amyloid who later developed full-blown amyloidosis [4, 6]. Lab 1 evaluated 14 rectal biopsies (unspecialized institutes of pathology using CR), lab 2 evaluated 18 renal biopsies (expert laboratory using EM), and lab 3 (reexamination of all 32 biopsies some years later) used CRIC (CR and immunohistochemistry double staining). Most of the rectal biopsies were early biopsies obtained at the beginning of the onset of systemic amyloidosis, and most of the renal biopsies were late biopsies. (Reproduced with written permission from Linke RP, Gärtner V, Michels H. High-sensitivity diagnosis of AA amyloidosis using Congo red and immunohistochemistry detects missed amyloid deposits. J Histochem Cytochem. 1995;43:863–9)
Fig. 14.4
High-sensitivity detection of early AA amyloid in children with rheumatoid arthritis, retrospective study using CRIC. This highly sensitive method identified extremely small amyloid deposits that were present in the early stages of AA amyloidosis that were originally missed at the time of biopsy (see text) since this CRIC method was developed only years later [4–6]. (a) Minute amyloid deposits of early amyloid detected by CRIC in the lamina propria of a rectal biopsy (dark brown dots). (b) Small glomerular amyloid deposits as detected by CRIC (low power, one glomerulus does not show amyloid). (c) The boxed area of (b) is magnified to better recognize the few amyloid spots and their site within the glomeruli (glomeruli are indicated by arrows). (Reproduced with written permission from Linke RP, Gärtner V, Michels H. High-sensitivity diagnosis of AA amyloidosis using Congo red and immunohistochemistry detects missed amyloid deposits. J Histochem Cytochem. 1995;43:863–9, modified)
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
