Gastrointestinal amyloid deposits in old age

Path. Res. Pract. 190,641-649 (1994)

Original Papers

Gastrointestinal Amyloid Deposits in Old Age Report on 110 Consecutive Autopsical Patients and 98 Retrospective Bioptic Specimens * C. R6cken 1,2, W. Saeger1 and R. P. Linke3 1Department of Pathology, Marienkrankenhaus, Hamburg; 21nstitute of Pathology of the University of Hamburg; 3Max-Planck-lnstitute of Biochemistry, Martinsried, Germany

SUMMARY

The prevalence of gastrointestinal amyloid was determined in 110 consecutive autopsy patients aged 85 years and older. Paraffin sections from the esophagus, stomach, small intestine, colon and rectum were stained with Congo red and inspected in polarized light. Amyloid was found in 38 patients (36%). Four patients had generalized amyloidosis and the remaining 34 revealed more localized varieties of amyloid. Immunohistochemical classification with a panel of antisera directed against five major amyloid fibril proteins displayed at least four different types of amyloid. Twenty cases exhibited amyloid of transthyretin origin, five cases were of lambda light chain origin and one case was of amyloid A-type. Thus far, unclassifiable amyloid deposits were found in 18 cases, 16 of which were strictly localized in subserosa I veins. This consecutive dutopsy series was compared with 98 randomly-selected bioptical gastrointestinal tissues with amyloid according to distribution pattern, sex, frequency of the various types of amyloid and associated diseases.

Introduction The prevalence of amyloid deposition of multiple origin increases with age 6,20,31,39. Apart from the more generalized, senile cardiovascular amyloidosis of transthyretin origin5,27, more organ-limited deposits affect different organs and tissue sites in old age, such as the heart (atrial natriuretic peptide 33 ), brain (A~ protein 9,24), pituitary gland 36 , jointslO, intervertebral discs 2o ,41, cartilage 25 , aorta 26,38 and seminal vesi-

* Results were presented in part at the 77th congress of the

Deutsche Gesellschaft fur Pathologie in Wiirzburg, Germany,

June 1993. _ Supported by the Johanna and Fritz Buch-Gedachtnisstiftung, Hamburg, Germany. © 1994 by Gustav Fischer Verlag, Stuttgart

cles 28 . Comprehensive data about gastrointestinal amyloid are not available, although senile cardiovascular amyloidosis has also been found in the gastrointestinal tract 27 . Affection of the gastrointestine in cases of generalized amyloidosis is a common feature with frequencies up to 97%7. Amyloid deposits with more restricted localization have also been described 2,12. According to the biochemical classification, all known systemic types of amyloid affect the gastrointestine, i.e. amyloid A (AA), amyloid of lambda (AA) and kappa (AK) light chain origin, or amyloid of transthyretin (ATTR) and ~2-microglobulin (A~2M) origin. Immunohistochemical studies of large autopsical and bioptical collectives proved AA, AA and AK to be classified most frequently corresponding to the different types of systemic amyloidosisll,13,22,23,37. Barreton et 0344-0338/94/0190-0641$3.50/0

642 . C. Rocken et al. Table 1. Antibodies used for immunohistochemical classification of amyloid Antibody

Animal

MonoclonaV polyclonal

Dilution

Method

Reference

anti-AA (mcl)l anti-Ale (HAR)2 anti-AK (SIN)2 anti-ATTR (TIE)2 anti-A~2M (WOE)2 anti-amyloid-P component

mouse rabbit rabbit rabbit rabbit rabbit

m p p p p p

1:10 1:1200 1:1000 1:800 1:200 1:800

PAP ABC ABC ABC ABC ABC

18, 19,21,22

unpublished

lCulture supernatant, code of monoclonal antibody; 2Initials of the patients from whom the amyloid fibrils were isolated. PAP =Peroxidase-antiperoxidase, ABC =Avidin-biotin-complex, AA = amyloid A, ATTR = amyloid of transthyretin origin, Ale = immunoglobulin lambda-light chain amyloid, AK = immunoglobulin kappa-light chain amyloid, A~2M = amyloid of ~2microglobulin origin, AX = amyloid of unknown origin. AN = amyloid undetermined.

al. l and Chastenoy et al. 3 evaluated an age-dependent distribution pattern of different types of generalized amyloidoses. However, these studies were performed on selected collectives with proven amyloidosis. A consecutive post-mortem series was not evaluated. More age-dependent amyloid deposits restricted to gastrointestinal amyloid have not been considered thus far. The aims of our study were to evaluate the prevalence and the distribution pattern of gastrointestinal amyloid deposition in patients aged 85 years and older. We attempted to determine the organ pattern of the esophagus, stomach, small intestine, colon and rectum. To identify the unknown types, immunohistochemical classification of amyloid was performed with a panel of five different amyloid-recognizing antisera. In addition, we compared these data with a bioptic series of amyloid-containing gastrointestinal tissues with special reference to the distribution pattern, different types of amyloid, differences of age and sex, and of the association to underlying disorders. Material and Methods Autopsical Specimens Samples of the esophagus, stomach, small intestine, colon and rectum were selected from 110 consecutive autopsies in patients aged 85 years and older, irrespective of disease or cause of death. Formalin-fixed (4% buffered formaldehyde in distilled water) and paraffin-embedded specimens were stained with Congo red 29 and inspected using a Leitz microscope in bright and polarized light with tension free optics. Evaluation of Congo red-stained sections was performed with reference to the localization of amyloid in the different layers of the gastrointestine (mucosa, submucosa, muscularis, adventitia/subserosa paraproctium) and to the tissue structure (vascular, interstitial). Finally, the number of afflicted vessels was counted in every specimen. All samples with amyloid underwent immunohistochemical analysis. Immunohistochemical studies were performed using the peroxidase-antiperoxidase (PAP) or the Avidin-Biotin (ABC) methods (Vector Laboratories, Burlingame, CA., USA). Deparaffinized sections were initially incubated with ". For abbreviations see Table 1.

the primary antibodies for 30 minutes at room temperature. Antibodies used, the dilutions and procedures were performed as listed in Table 1. The reaction was visualized with 3,3-diamino-benzidine-tetrahydrochloride (DAB). The counterstain was hematoxylin. Positive controls were performed with different specimens containing known types of amyloid, including AA, Ale, AK, ATTR and A~2M*. Another control was performed on serial sections, whereby areas stained with Congo red and corresponding immunohistochemically stained areas were compared. Specificity controls included omission of the first antibody and absorption of the specific antibody with the respective antigen. The reactions were consistent with the assumption of specificity.

Bioptical Specimens From 1974 to 1992, all bioptical specimens for this study with amyloid of the gastrointestine were selected from a general hospital. Age, sex, localization of biopsy, histochemical and immunohistochemical classification of amyloid were registered. All bioptical specimens had been stained with Congo red, some with and some without previous oxidation as described by Wright et a1. 40 • As far as immunohistochemical classification was concerned, the procedure has been the same as described above.

Results Autopsical Specimens Thirty-eight (35 %) of 11 0 autopsied patients aged 85 years and older had amyloid deposits in the gastrointestine. The mean age of patients with amyloid was 88.84 years (range 85-99 years) with no significant differences when compared with the whole· series of 110 patients (mean age 87.9; range 85-99 years). The mean age of men (87.8 years) and women (89.0 years) with amyloid deposits of the gastrointestine did not reveal significant differences. While the sex ratio of the whole autopsic series was 2.5:1 (female to male), the prevalence of gastrointestinal amyloid deposits in these groups was 6.6:1. This is a 2.64-fold increase of amyloid deposits in women (p-value < 0.02). Five different gastrointestinal localizations were studied in each patient, i.e. esophagus, stomach, small

Gastrointestinal Amyloid Deposits in Olde Age· 643 25

.

I_ Vascular

o Interstitiall

Esophagus

Stomach

22

22

20

=

I.

c.

15

~

t ~

.c 10

Z

5

o Fig. 1. Distribution of amyloid deposits at different sites in autopsic specimens.

intestine, colon and rectum, reaching a total number of 550 specimens in the whole autopsic series. Amyloid was found in 83 of these 550 specimens. The rectum (24 patients) and colon (22) were affected most commonly. Deposits in the small intestine (16), esophagus (11) and stomach (10) were less frequent. Affection of a single localization was identified in 18 patients. Amyloid deposits at two, three and four different localizations were observed in 14 patients. Amyloid deposits at all localizations, i.e. esophagus, stomach, small intestine, colon and rectum, were displayed in six of 38 cases. Four of these revealed generalized amyloidosis. Generalized amyloidosis without participation of the gastrointestine or with affection of less than five different localizations was not observed tn the autopsical series, whereas amyloid deposits at all localiza18

Intestine Site

tions without generalized amyloidosis were displayed in two cases. The number of specimens with deposits in the submucosa and subserosa was similar, each being 55. Twenty-one specimens included deposits in the tunica muscularis. Only seven specimens exhibited amyloid in the mucosa and five revealed deposits in the muscularis mucosae. In the whole autopsical series, the frequency of vascular deposits far exceeded interstitial deposits (Fig. 1). Meanwhile, individual cases failed to reveal both vascular and interstitial affection. One patient exhibited corpora amylacea in the mucosa of the stomach and a second patient proved to have intracellular, congophilie, fibrillar inclusions in the intestinal mucosa.

10 ATTRO AA

16

Rectum

Colon

• AA

rn

AX

14

:l 12

Ii

!.10 t

..Q

8

E

=

Z

Fig. 3. Tissue distribution of gastrointestinal amyloid in autopsic specimens of 36 patients, (Abbreviations see Table 1). *Subserosal layer refers to adventitia (esophagus), subserosa (stomach, intestine and colon) and paraproctium (rectum).

6

Mucosa

Submucosa

Muscularis

Subserosa"

16

644 . C. Rocken et al. 60

. 1:1 QI

~.=-

OS

I_ Vascular

o Interstitiall

55

50 40 30

t

; ;Z;

A

20

10 0 Stomach

Colon

Intestine

Rectum

Site

Classification of amyloid deposits was performed with Congo red and immunohistochemistry on serial sections. The amyloid deposits of nine spe<:imens remained unclassified (Table 2). The amyloid deposits were too small and sparse in these cases and could not be localized with certainty in the immunhistochemically stained serial sections. Immunohistochemical classification of amyloid in the gastrointestine revealed at least four different types of amyloid in the remaining 35 patients (Table 2). Twenty cases exhibited amyloid of transthyretin origin (Fig. 2), five cases were of lambda light chain origin and one case was amyloid A. Amyloid of kappa light chain and ~2-microglobulin origin were not identified. Undassifiable were the amyloid deposits in 18 cases (Table 2), including two cases with mucosal deposits, one with corpora amylacea and another with intracellular, fibrillar inclusions. The cell type with intracellular inclusions could not be identified because of advanced autolysis. Sixteen unclassifiable cases demonstrated amyloid deposits in subserosal vessels alone. Serial sections with Congo red and van Gieson's elastica staining verified these deposits as being localized in subserosal veins in every case except one where deposits were found in both the submucosal and subserosa I veins. In eight out of sixteen cases, the unclassifiable deposits exhibited an immunoreactivity with anti-amyloid P-component. The immunoreaction was patchy and did not enclose the entire amyloid deposits. The four patients with gastrointest-

Fig. 5. Distribution of amyloid deposits at different sites in bioptic specimens.

ina 1 involvement in systemic non-heriditary amyloidosis were identified as having senile cardiovascular amyloidosis of the transthyretin origin. The tissue distribution of the different amyloid types were differed markedly; transthyretin amyloid was found mainly in the submucosa, less between muscle layers and only sparsely in the subserosa. Three cases with transthyretin-derived amyloid exhibited deposits only in the subserosa. Unclassifiable vascular deposits were almost exclusively in the subserosa (Fig. 3). Co-localizaton of different immunoreactivity in the gastrointestine was revealed in eight patients (Table 3). Three cases exhibited transthyretin and unclassifiable deposits in different vessels, the former in arteries of the subserosa and submucosa, and the latter strictly in subserosal veins. Two patients revealed immunoreactivity with anti-AI" and anti-ATTR in the same vessels, while two cases exhibited AA- and ATTRimmunoreaction at the same site in different vessels. One patient with generalized senile cardiovascular amyloidosis of transthyretin origin additionally exhibited local deposits of AA- and AA-immunoreactivity both in the same (AA and ATTR, ATTR and AI,,) and in different vessels (AA and ATTR - Fig. 4). No case of generalized AA- or AA-amyloidosis was found. Comparison of the autopsically diagnosed diseases of the amyloid-type failed to reveal any correlation to the underlying diseases (Table 2) since no disease was represented more frequently in patients with gastrointestinal amyloid deposits as compared with the

Fig. 2. Uneffected immunoreaction in advanced autolysis of the intestine. Congo red staining (a) and immunohistochemistry (b), anti-ATTR, ABC, hematoxylin; original magnification 170-fold: - Fig. 4. Local vascular amyloid deposits of AA-type in senile cardiovascular amyloidosis of transthyretin origin. Congo red staining (a) and immunohistochemistry: Anti-AA (b), antiATIR (c), ABC, hematoxylin; original magnification 310-fold. - Fig. 6. Vascular and interstitial amyloid deposits in the mucosa, muscularis mucosae and submucosa of rectum biopsy. Congo red staining (a) and immunohistochemistry (b), anti-AA, PAP, hematoxylin; original magnification 170-fold.

~

646 . C. Rocken et al. Table 2. Correlation of amyloid-type and clinical and autopsic diagnosis

Clinical and autopsical diagnosis

Bioptic series UnAmyloid class 1 deterImmunohistochemistry with antibodies against mined AA AK ATTR AA n

Rheumatoid arthritis Nephrotic syndrome Chronic hepatitis B Familial Mediterranean fever Spondylitis ankylosans Asbestosis Whipple's disease Crohn's disease Necrotizing duodenitis Retroperitoneal fibrosis Malignant tumor

n

n

n

24 3 1 1 4 1 1 1

Myelodysplastic syndrome Myeloproliferative syndrome Monoclonal gammopathy Plasmacytoma Chronic lymphatic leukemia

n

4 40 44

n

n

n

2

1

1

7

7

8

1

21

1 1

1

Non-infectious and nonmalignant disease

No clinical data Total number of patients Number of specime11s

n

UnPatients deter- without mined gastrointestinal amyloid n n 1

2 4

Tumor-like lesion in duodenum Idiopathic amyloidosis Familial amyloidosis

Autopsic series Amyloid class 1 Immunohistochemistry with antibodies against ATTR AX AA AA

1 2 13

23 38

1

1 1 1

1

1 1

10

6

3

1

2

41

20 51

18 27

5 7

1 3

3b 9h

72 467

2 3 3 5

2 2 4

5 7a 7a

1Abbreviations see Table 1. - aThese cases remained undetermined because the whole panel of antisera had not been available in earlier years or the sample size was too small for further immunohistochemical evaluation. - bNo amyloid in second Congo red stained serial section for immunohistocherp.ical classification.

whole autopsic series. Clinical symptoms due to amyloid deposition in the gastrointestine were available in only one patient with senile cardiovascular amyloidosis who suffered from chronic diarrhea. Other infectious and non-infectious diseases associated with diarrhea were excluded in this case. Bioptical Specimens Ninety-eight biopsies of 75 patients (41 women, 34 men) with amyloid deposits in the gastrointestine were available for analysis. The mean age of the collective was 60.3 years with a range from 19 to 93 years. The mean age of men (59.8 years) and women (60.7 years) did not reveal significant differences. The patients with amyloid had a female to male ratio of 1.2:1, although this prevalence was not significant. Sixty-four bioptic specimens were from the rectum, 15 from the small intestine, 14 from the stomach and five from the colon. Biopsies were obtained from only one localization in 64 patients. Biopsies at two or three

different localizations were obtained from 11 patients. Biopsies of five different localizations, i.e. esophagus, stomach, small intestine, colon and rectum, were not available from any of the patients. Congo red staining revealed amyloid deposits most commonly in the submucosa (92 specimens), and less frequently in the mucosa (38) and muscularis mucosae (30). Tunica muscularis or subserosal layers were not included in any biopsy. Vascular and interstitial deposits were found in every layer (Fig. 5). In the rectum, vascular deposits were more frequent (55 specimens) than interstitial deposits (34 specimens). At the other sites, i.e. stomach, intestine and colon, the number of specimens with vascular and interstitial deposits was nearly equal (Fig. 5). Immunohistochemical classification yielded AAamyloid in 40 cases (Fig. 6), AA-amyloid in 23 cases, AK-amyloid in three cases and two cases of transthyretin origin (Table 2). Seven biopsies in seven cases could not be classified (Table 2). Two of these patients were unclassified because the whole panel of antisera

Gastrointestinal Amyloid Deposits in Olde Age· 647 Table 3. Autopsical patients with multiple types of amyloid Amyloid ATTR ATTR ATTR ATTR ATTR

+ AX +A'A + A'A + AA + A'A + AA

Number of patients

}

Total

3 4

1 8

{

Number of sites 4 4 2 2 1 13

lAbbreviations see Table 1.

had not been available in earlier years and no specimens were available for reexamination at the end of the study. Immunohistochemical evaluation was performed with anti-AA and revealed a negative immunoreactivity. The remaining five cases were undetermined because the sample size had been too small for further immunohistochemical evaluation. Comparison of clinical diagnosis with immunohistochemical classification revealed a very strong correlation between the underlying disease and the amyloid type found. The results are summarized in Table 2.

Discussion Thirty-eight (35%) patients of a consecutive and unselected autopsic series of 110 patients aged 85 years and older revealed amyloid deposits in the gastrointestinal tract. This is an unexpectedly high prevalence. Immunohistochemical classification revealed at least four different types of amyloid affecting the gastrointestine in old age: amyloid of transthyretin origin, amyloid A, amyloid ",-light chain and, as, yet unclassified deposits. Twenty patients exhibited amyloid of transthyretin origin. Deposits of transthyretin amyloid were primarily localized in the submucosal layer (Fig. 3), with decreasing frequency into outer layers, and three patients exhibited ATTR only in the subserosa. Four cases revealed ATTR amyloid of the gastrointestine in generalized senile amyloidosis of transthyretin origin. This amyloid type is not coupled with a known underlying disease but may instead lead to such severe complications as congestive heart failure 32 . ATTR deposits of the remaining patients, even those two cases with deposits at all localizations without generalized amyloidosis, are interpreted as local spots of a systemic disorder which might have later developed into a disease with the affection of multiple organs and tissues. In the entire autopsic series, immunohistochemical evaluation was performed with serial sections, one of which was stained by Congo red. Comparison of Congo red and immunohistochemically stained sections revealed co-localization of A", and ATTR deposits in the same and in different vessels in five cases. A", and

ATTR in the same vessels was observed in three cases and is considered to be an adsorption of one of the proteins, as previous analysis has shown 22 • Amyloid", deposits in ATTR-negative vessels and in two ATTRpositive cases are considered to be local deposits without known myeloproliferative disease. At least one of these two patients with generalized senile cardiovascular amyloidosis and local A'k deposits exhibited additional local immunoreactivity with anti-AA, both in different vessels and in the vessels which were already affected by ATTR-deposits. This 93-year-old women did not suffer from a known chronic infectious disease nor a known myeloproliferative disorder. Extensive immunohistochemical studies of various organs and tissue sites failed to reveal generalized AA or A", deposits 30 , and the pathogenesis of these deposits remains uncertain. The identification of more local AA-derived amyloid deposits is not recognized so far, except in dogs 4 • Immunohistochemical classification of gastrointestinal amyloid deposits in the autopsic series exhibited a fourth, still unclassified, precursor-protein which was almost exclusively localized in subserosa I veins except in one patient with non-reactive deposits in both the submucosal and subserosal veins. Comparison of the autopsic series with a bioptic series of our general hospital reveals differences with regard to age, sex and the distribution pattern of the various types of amyloid. The consecutive autopsic series demonstrated a mean age of 87.9 years, whereas the bioptic series revealed a men age of 60.3 years. The prevalence of amyloid deposition is increased in women in the autopsical series, while the sex ratio in the bioptical series is nearly equal to one. Indeed, the sex ratio of the whole autopsical series is in favor of women according to the mortality of the different sexes, although the mean age of mean and women fails to reveal significant differences. The increased prevalence of amyloid in women is thus interpreted as being sex-specific. Different authors describe an age-dependent distribution pattern of the various types of systemic amyloidosis 3 , in that AA and A", are more common in the younger series of bioptical specimens, whereas amyloid of transthyretin origin is more frequent in old age. Our data confirm this observation. Furthermore, amyloid deposits in bioptical series demonstrate a correlation to an underlying disease, whereas amyloid deposits in the autopsical series do not. This difference is due to the selection procedure. The bioptical series is a collective which is strictly selected by the presence of amyloid and consists of samples which were obtained partly with special request by clinicians for the histopathologic examination of amyloidosis. These patients suffered from diseases which may lead to the formation of amyloid, for example rheumatoid arthritis or myeloproliferative diseases not strictly coupled with advanced age. The bioptical series reflects the distribution pattern of the various types of systemic amyloidosis with a mean age younger than that of the

648 . C. Rocken et al.

autopsical collective 11,13,15,16,22. Bioptical specimens of Ale-immunoreactive, tumor-like deposits restricted to one organ (duodenum) were considered to be local or organ-limited deposits. The autopsical series is a contino us sample irrespective of diseases and cause of death. Selection was performed only by age. Thus, the amyloid deposition in autopsical specimens reveals the true prevalence of gastrointestinal amyloid in old age. Amyloid is biochemically classified according to the various precursor proteins. Each protein is associated with different disorders and clinical settings. The deposition pattern of amyloid is divided into generalized, organ-limited and local affections 8,14,34. Endoscopically obtained biopsies of the gastrointestine are routine procedure in the diagnosis of amyloidosis. While Congo red staining and polarizationmicroscopy verifies the presence of amyloid in the gastrointestine, the subsequent immunohistochemical differentiation of the amyloid deposits reveals the amyloid class and identifies the precursor protein. Since there is a strong correlation of clinical syndrome and deposited protein 8 , it affirms the clinical data of an underlying disease 27,37. Histopathologic diagnosis of amyloid in bioptic gastrointestinal specimens could, by itself, not be used to differentiate whether the deposits are part of generalized amyloidosis, local or organ-limited amyloid deposits. Further clinical investigations, endoscopicaf35 and radiologic findings!7, as well as patients history, may ~be helpful in diagnosis, while knowledge about the prevalence and distribution pattern of different types of amyloid in the gastrointestine are necessary to evaluate the deposition pattern and underlying disease. Routine bioptical samples exclude deep layers like the muscularis and subserosa. Diagnosis of amyloid is feasibk if the sample covers enough material and such histoanatomical structures as vessels. The most common types of amyloid are detectable under these circumstances, even amyloid of old age, i.e. amyloid of transthyretin origin (Fig. 3). However, our data show that some amyloid deposits will be missed in routine biopsies. Three patients with ATTR and 15 with unclassified deposits exhibited amyloid in deep layers and could not to be identified in biopsy (a total of 47% of the whole amyloid-positive autopsic group). The differential diagnosis is further complicated by the possibility of the presence of more than one type of amyloid per specimen or organ, especially in old age 30 . In case of "multiple amyloid" disease, the deposition pattern and clinical relevance of each type could be estimated if biopsies are obtained from different localizations in the gastrointestine or from various organs, such as the gastrointestine, liver and kidney. Many publications revealed an increased prevalence of amyloid deposition reaching a frequency of up to 80% in the eighth decade of life (for review see 2). These deposits are interpreted as being local and organ-limited amyloid of multiple origin affecting different organs. Sixteen patients exhibited unclassifiable

amyloid deposits restricted to subserosal gastrointestinal veins. Future investigations will have to examine whether these deposits are more local or more organlimited, whether they are age-dependent deposits or correlate to some disease. Concluding our data, gastrointestinal amyloid deposits of different chemical types may appear in old age, including AA, ATTR, Ale and a hitherto unclassified amyloid. Gastrointestinal amyloid was exhibited in 35% of autopsical patients studied in our series. 50% of these cases proved to be deposits of transthyretin origin, 44% of unknown origin and 6% with both transthyretin and unclassified amyloid, likewise in nongeneralized affection.

References 1 Baretton G, Linke RP, Lohrs U (1990) Systemische Amyloidosen. Immunhistochemische Typisierung an Autopsien mit Hilfe zahlreicher spezifischer Antikorper. Pathologe 11: 71-79 2 Bethke BA, Stein au G, Schubert GE, Burger L (1983) Isolierte tumorformige Amyloidose des Magens. Med Welt 34: 1422-1425 3 Chastonay P, Hurlimann J (1986) Characterization of different amyloids with immunological techniques. Path Res Pract 181: 657-663 4 Colbatzky F, Brunnberg L, Linke RP, Geisel 0, Hermanns W (1991) AA-like amyloid deposits confined to arthritic joints in two dogs with rheumatoid arthritis. J Comp Patholl05: 331-343 5 Cornwell GG, III., Murdoch W, Kyle RA, Westermark P, Pitkanen P (1983) Frequency and distribution of senile cardiovascular amyloid. A clinicopathologic correlation. Am J Med 75: 618-623 6 Cornwell GG, III., Westermark P (1980) Senile amyloidosis: a protean manifestation of the aging process. J Clin Pathol 33: 1146-1152 7 Gilat T, Revach M, Sohar E (1969) Deposition of amyloid in the gastrointestinal tract. Gut 10: 98-104 8 Glenner GG (1980) Amyloid deposits and amyloidosis. New Engl J Med 302: 1333-1343 9 Glenner GG, Wong CW (1984) Alzheimer's disease: initial report of the purification and characterization of a novel cerebrovascular amyloid protein. Biochem Biophys Res Commun 120: 885-890 10 Goffin YA, McCrickard EL, Ameryckx JP, Malmendier CL, Hiden M, Cornwell GG, III. (1985) Amyloidosis of the joints: evidence that human hip capsules have a unique predisposition for amyloid of the senile systemic type. Appl Pathol 3: 88-95 11 Hoshi Y, Iwata T, Kawano H, Takahashi M, Ishihara T, Uchino F (1992) Immunohistochemical classification of 109 autopsy cases of systemic amyloidosis in Japan. Patologia 25: 212-213 12 Ikeda K, Murayama H (1978) A case of amyloid tumor of the stomach Endoscopy 10: 54-58 13 Iwata T, Yokota T, Ishihara T, Takahashi M, Yamashita Y, Nagasawa T, Gondo T, Kawano H, Kamei T, Okuzono Y, Uchino F (1988) Histologic characterization of amyloidosis based on amyloid proteins. In: Isobe T, Araki S, Uchino F, Kito S, Tsubura E (eds) Amyloid and Amyloidosis, Plenum Publishing Corporation, New York, pp 795 -799

Gastrointestinal Amyloid Deposits in Olde Age . 649 14 Kisilevsky R (1991) Amyloid and amyloidoses: differences, common themes, and practical considerations. Mod Pathol 4: 514-518 15 Kyle RA, Bayrd ED (1975) Amyloidosis: review of 236 cases. Medicine 54: 271-299 16 Ky1e RA, Greipp pR (1983) Amyloidosis (AL). Clinical and laboratory features in 229 cases. Mayo Clin Proc 58: 665 -683 17 Legge DA, Carlson HC, Wollaeger EE (1970) Roentgenologic appearance of systemic amyloidosis involving gastrointestinal tract. Am ] Roentgenol Rad Therapy Nucl Med 110: 406-412 18 Linke RP (1984) Monoclonal antibodies against amyloid fibril protein AA. Production specificity, and use for immunohistochemicallocalization and classification of AA-type amyloidosis. ] Histochem Cytochem 32: 322-328 19 Linke RP (1985) Immunochemical typing of amyloid deposits after microextraction from biopsies. Appl Pathol 3: 18-28 20 Linke RP (1989) Amyloid deposits derived from various proteins - a hallmark of old age. In: Platt D (ed) Gerontology, Springer, Berlin, Heidelberg, pp 175 -192 21 Linke RP, Geisel 0, Mann K (1991) Equine cutaneous amyloidosis derived from an immunoglobulin lambda-light chain. Immunohistochemical, immunochemical and chemical results. Bioi Chern Hoppe-Seyler 372: 835 - 843 22 Linke RP, Nathrath WB], Eulitz M (1986) Classification of amyloid syndromes from tissue sections using antibodies against various amyloid fibril proteins: report of 142 cases. In: Glenner GG, Osserman EF, Benditt EP, Calkins E, Cohen AS, Zucker-Franklin D (eds) Amyloidosis, Plenum Publishing Corporation, New York,~pp 599-605 23 Looi LM, Sumithran E (1988) Morphologic differences in the pattern of liver infiltration between systemic AL and AA amyloidosis. Hum Pathol19: 732-735 24 Masters CL, Simms G, Weinman NA, Multhaup G, McDonald BL, Beyreuther K 85 (1985) Amyloid plaque core protein in Alzheimer disease and Down syndrome. Proc Nat! Acad Sci USA 82: 4245 -4249 25 Mohr W, Kuhn C, Linke RP, Wessinghage D (1991) Deposition of amyloid of unknown origin in articular cartilage. Virchows Archiv [Bj 60: 259-262 26 Mucchiano G, Cornwell GG, III., Westermark P (1992) Senile aortic amyloid. Evidence for two distinct forms of localized deposits. Am] Pathol 140: 871-877 27 Pitki"inen P, Westermark P, Cornwell GG, III. (1984) Senile systemic amyloidosis. Am] Pathol117: 391-399 28 Pitkanen P, Westermark P, Cornwell GG, III., Murdoch W (1983) Amyloid of the seminal vesicles. A distinctive and

common localized form of senile amyloidosis. Am ] Pathol 110: 64-69 29 Pucht!er H, Sweat F, Levine M (1962) On the binding of Congo red by amyloid.] Histochem Cytochem 10: 355-364 30 Rocken C, Saeger W, Linke RP (1993) Mehrere unterschiedliche Amyloidtypen bei einer 93-jahrigen Patientin: Kasuistik eines Sektionsfalles. Pathologe 14: 42-46 31 Schwartz PH (1970) Amyloidosis. Cause and manifestation of senile detoriation. c.c. Thomas, Springfield, Illinois, USA 32 Smith T], Kyle RA, Lie]T (1984) Clinical significance of histopathologic patterns of cardiac amyloidosis. Mayo Clin Proc 59: 547-555 33 Steiner I (1987) The prevalence of isolated atrial amyloid. ] Pathol 153: 395 -398 34 Stone M] (1990) Amyloidosis: a final common pathway for protein deposition in tissues. Blood 75: 531: 545 35 Tada S, Iida M, Iwashita A, Matsui T, Fuchigami T, Yamamoto T, Yao T, Fujishima M (1990) Endoscopic and biopsy findings of the upper digestive tract in patients with amyloidosis. Gastrointest Endosc 36: 10-14 36 Tashima T, Kitamoto T, Tateishi], Ogomori K, Nakagaki H (1988) Incidence and characterization of age related amyloid deposits in the human anterior pituitary gland. Virchows Archiv [Aj 412: 323-327 37 van de Kaa CA, HoI PR, Huber], Linke RP, Kooiker C], Gruys E (1986) Diagnosis of the type of amyloid in paraffin wax embedded tissue sections using antisera against human and animal amyloid proteins. Virchows Archiv [Aj 408: 649-664 38 Wright ]R, Calkins E (1974) Relationship of amyloid deposits in the human aorta to aortic atherosclerosis. A postmortem study of 100 individuals over 60 years of age. Lab Invest 30: 767-773 39 Wright ]R, Calkins E, Breen W], Stolte G, Schultz RT (1969) Relationship of amyloid to aging. Review of the literature and systemic study of 83 patients derived from a general hospital population. Medicine 48: 39-60 40 Wright ]R, Calkins E, Humphrey RL (1977) Potassium permanganate reaction in amyloidosis. A histologic method to assist in differentiating forms of this disease. Lab Invest 36: 274-281 41 Wullbrand A, Saeger W, Missmahl HP, Voigt C, Linke RP (1990) Amyloid in intervertebral discs of surgery and autopsy material. A new class of amyloid? Virchows Archiv [Aj 416: 335-341

Received March 23, 1994 . Accepted April 22, 1994

Key words: Gastrointestine - Amyloid - Immunohistochemistry Prof. Dr. W. Saeger, Department of Pathology, Marienkrankenhaus, AlfredstraBe 9, D-22087 Hamburg, Germany, Tel.: 40/2546-2471, FAX: 40/258885