close

Вход

Забыли?

вход по аккаунту

?

Giant cell inflammation compared with amyloidosis of the internal elastic lamina in temporal arteries.

код для вставкиСкачать
1186
BRIEF REPORT
GIANT CELL INFLAMMATION COMPARED WITH AMYLOIDOSIS OF
THE INTERNAL ELASTIC LAMINA IN TEMPORAL ARTERIES
THOS. J. MUCKLE
Close examination of a series of temporal artery
biopsy specimens provided histologic and statistical evidence that any association between the occurrence of
giant cell arteritis and senile amyloidosis of the temporal
artery internal elastic lamina is not direct, but is associated incidentally with old age. Substantial demographic differences between the 2 conditions were
found; the prevalence of giant cell arteritis first waxed
and then waned with increasing age, while rates for
senile amyloidosis of the temporal artery progressed
steadily toward 100% in the ninth decade of life.
Both giant cell arteritis-polymyalgia rheumatica syndrome (GCA-PMR) and senile amyloidosis of
the internal elastic lamina of the temporal artery
(SATA) occur largely in the elderly. The pathogenesis
of neither condition is understood, but their agerelated concurrence might indicate a causal relationship. This idea was supported by the findings of a 1975
study in which all 10 temporal arteritis biopsy specimens showed apparent amyloid transformation of the
internal elastic lamina (IEL) (1).
Any other connection between these 2 diseases
From the Department of Pathology, McMaster University,
and the Department of Laboratory Medicine, Chedoke-McMaster
Hospitals, Hamilton, Ontario, Canada.
Supported in part by the Canadian Medical Research and
Rheumatism Research Councils and the Wilson Trust Fund of
Chedoke Hospitals.
Thos J. Muckle, MD, FRCP(C): Director of Laboratories,
Chedoke Division, Chedoke-McMaster Hospitals.
Address reprint requests to Thos. J . Muckle, MD,
FRCP(C), Director of Laboratories, Chedoke Hospital Division,
Chedoke-McMaster Hospitals, PO Box 2000, Station A, Hamilton,
Ontario L8N 325, Canada.
Submitted for publication September 1 , 1987; accepted in
revised form March 1, 1988.
Arthritis and Rheumatism, Vol. 31, No. 9 (September 1988)
is not evident in the extensive literatures of either
amyloidosis or GCA-PMR, although GCA-PMR with
cerebral amyloid angiopathy has been described ( 2 4 )
and several examples of GCA-PMR with systemic
amyloidosis are on record (5-8). Both GCA-PMR
(9,lO) and SATA (1 1) are not uncommon in the senium, and this fact is strange in contrast with the
isolation of the single report of their coincidence (1).
Therefore, a review of a group of temporal artery
biopsy specimens was undertaken in the hope of
defining more closely any relationship between GCAPMR and SATA. The findings suggest that the connection is incidental, and that both are independently
associated with old age, rather than with each other.
PATIENTS AND METHODS
Multiple histologic sections of routinely processed surgical biopsy specimens of temporal artery
from 122 white patients with suspected giant cell
arteritis were reviewed.
Stains used included hematoxylin and eosin,
saline-alkaline Congo red (including checking for permanganate resistance) (12,13), Sirius red (14), Leung’s
amyloid-elastic stain ( 1 9 , standard toluidine blue (16),
Weigert’s and Verhoeff-van Gieson’s elastic tissue
stains, diastase periodic acid-Schiff, acetic acid-Alcian blue, and the Adams technique (17) for tryptophane. Regular controls were included in every group.
Scanning for the presence and grade of both
diseases was carried out in tungsten light. Ultraviolet
light inspection of Congo red stain sections was used
as well, for amyloid scanning.
The severity of active GCA was expressed as 5
grades, from 1 (normal, including senile changes) to 5
BRIEF REPORTS
(round cells and multinucleated giant cells so dense as
to substantially obscure mural architecture, with more
than 75% circumference wall damage). In this grading
system, no account was taken of fibrinoid necrosis,
fragmentation of the IEL, or thrombosis, since all of
these may be associated with other conditions. Neither fibrous mural scars nor organized thrombosis
without active inflammation were counted as arteritis.
The extent of SATA was graded on a scale of
1-6, where 1 = no trace of amyloid visible in IEL
detectable with any form of illumination and inspection even under oil immersion, and 6 = numerous
heavy specks, streaks, bands, caps, and nodules of
amyloid throughout the circumference, amyloid transformation of the IEL usually present as well.
The series of sections was scored “blindly” on
separate occasions, several months apart, thrice for
prevalence and twice for grades, and the data generated were used for most computations without statistical transformation. Prevalence rates of presentation,
however, were modified for derivation of standardized
rates, and for graphic representation by smoothing and
fitting techniques, including annualization to detect
and compensate for possible distortion due to small
sample sizes (18,19).
RESULTS
Histologic observations. No differences in severity or detail of the patterns of amyloid deposition on
the IEL were detected in relation to the presence or
absence of GCA.
Observational consistency. There were no misclassifications for presence of GCA, whereas for
SATA, 3 discordant scores were obtained (in 3 runs)
when “absence” was scored as “trace” once, and
vice versa twice, yielding a K statistic (20) of 0.97
(standard error 0.05) and a ratio z of 18.5. There were
no discrepancies in the grading of GCA, but for SATA,
scoring inconsistencies between grades 4 and 5 occurred 4 times (in 122 assessments x 2), indicating
perhaps excessive grading subdivisions, but giving a
coefficient of stability (21) of 0.99, 2P < 0.001. The
number of subjects in each grade for both conditions is
given in Table 1.
Age, sex, and presence or absence of disease.
GCA affected both sexes equally, the male predominance in SATA was not statistically significant and
overall, GCA was less common than SATA (21%
versus 57%, chi-square [22] and t [23,24] 2F’ < 0.001).
There was no significant age difference between GCA-
1187
Table 1. Sex and disease grade of 122 patients with giant cell
arteritis (GCA) andlor senile amyloidosis of the temporal artery
(SATA)*
SATA
GCA
Grade
1 (normal)
2
3
4
415
5
6
Men
(n = 49)
Women
(n =73)
38
58
6
3
4
4
3
0
0
NA
4
0
2
NA
Men
(n = 49)
Women
(n =73)
18
9
11
3
2
5
35
6
11
13
2
3
3
1
* See Patients and Methods for explanation of grading system.
Grade 415 = graded 4 in one run, but 5 in the second. NA = not
applicable.
positive and GCA-negative groups (mean age If: SD
69.1 ? 1.8 versus 67.7 1.9 in men; 66.5 f 1.9 versus
68.0
1.4 in women), but both male and female
subjects with SATA were older, by 7 and 10 years,
respectively, than those without SATA (71.0 2 1.7
versus 63.5 5 2.5 in men, 2P < 0.02; 72.5 -+ 1.3 versus
62.6 f 1.7 in women, 2P < 0.001). In neither condition
did males and females differ in mean age.
The patterns of age-related incidence and prevalence differed markedly (2P < 0.001) between GCA
and SATA. The most outstanding feature in the incidence curves was that cases of GCA ceased to appear
beyond about the end of the eighth decade, while
examples of SATA continued to occur virtually to the
limit of the human life span. This suggests that GCA is
a time-limited condition, while SATA is permanent
(Figure 1). Twofold 77 coefficient values (25) for decade
annualized-decade prevalence rates were, for y on x
(predicting score from age), 0.98 for GCA and 0.99 for
SATA. For x on y (predicting age from score), the
values were 0.59 for GCA (resulting from the reflex
nature of the curve and the larger variance between
successive age rates) versus 0.99 for SATA. All 77
values were 2P < 0.001. Other correlation coefficients
(26) computed slightly lower, e.g., for GCA (quadratic) r = 0.89, 2P < 0.001, and for SATA (inverse
exponential) r = 0.98, 2P < 0.001. That the annualization procedure had not materially perturbed the
correlation results is demonstrated by simple linear
correlation coefficients for raw hemidecade rates of
0.89 and 0.95 for the ascending and descending limbs
of the GCA curve, and 0.96 for that of SATA.
Age and grade of disease. The relationship between age and grade of disease differed between the 2
conditions. Thus, mean decade age-group grade, mean
*
*
BRIEF REPORTS
1188
100
DISCUSSION
60
CT
n
40
50
60
AGE
70
80
90
100
IN YEARS
Figure 1. Biopsy-positive presentation prevalence (PREV) rates for
122 patients with giant cell arteritis (GCA) and/or senile amyloidosis
of the temporal artery (SATA). all subjects together. Curves are
constructed from regression data computed from annualized hemidecade rates.
age of grade groups, and correlation of individual ages
and grades, all for the whole group of subjects as well
as computed for affected-only, showed no (or statistically nonsignificant negative) association for GCA, but
a strong increase in grade with age for SATA. For
example, the correlation coefficients for SATA were,
respectively, r = 0.99 and 0.98,0.94 and 0.92, and 0.57
and 0.41 (all 2P < 0.001). The lower values for
individual age versus grade of SATA reflect the wide
age range of the subjects (41-96 years), which yields a
relatively large statistical variance, exclusion of the
prevalence effect in the affected-only groups, loss by
mortality attrition of older subjects with more advanced SATA, and individually variable ages at onset
and rates of progression of the condition.
Subjects by diagnosis and grade. There appeared
to be no significant association, by chi-square, between the occurrence of GCA and the occurrence of
SATA, for either sex. Nor was a correlation evident
between the respective grades of GCA and SATA (r =
0.13, 2P > 0.05). Finally, the grade means of SATA in
GCA-positive and GCA-negative groups, and the
grade means of GCA in SATA-positive and SATAnegative groups were compared and found to be
statistically indistinguishable (largest r = 0.54, 2P >
0.05).
In this study, GCA and SATA were found to be
associated in only 2 ways: by involving essentially the
same structure, the IEL of temporal arteries, and a
strong, albeit very different, demographic relationship
with advancing age. N o other connections between the
2 processes were found, but instead, there was considerable evidence of autonomy.
Thus, the patterns of incidence and prevalence
differ markedly. While both begin to appear around the
age of 50 and their rates increase rapidly with age up to
the late 60s, thereafter, rates for GCA diminish, while
those for SATA increase steadily to reach 100% in the
ninth decade. Again, no correlation between grade of
severity and increasing age was found with GCA,
while with SATA, the correlation was of a high order
and very similar to the prevalence characteristics.
Finally, no particular histologic or statistical association was demonstrated between the 2 processes.
Several of the specimens in a previously published report showed marked SATA but only early
GCA, giving rise to the suggestion that the amyloid
might have played a primary pathogenetic role in the
development of GCA ( I ) . These findings do not support that contention, but suggest rather that the association is merely coincidental to location and old age.
Nothing found in the present study excludes the
possibility of such a special pathogenetic connection,
but the data indicate that amyloid is at least unlikely as
a cause of GCA in more than a relatively small
proportion of cases.
Some reservations may be expressed in relating
the present data to patterns of natural history, since
the patient population was selected from the general
population on the basis not only of symptomatology ,
but also the clinical decision to perform a biopsy.
Information relevant to the present data is not available for symptomatic patients who were not biopsied,
nor symptom-free members of the base population.
Because of the circumstances of selection, it is not
possible to be sure how far the data recorded here may
misrepresent the true patterns of population ageprevalence of GCA and SATA. Insofar as SATA
appears to be asymptomatic, subjects may be considered essentially unselected, but subsidiary selection
may have occurred if an unknown association exists
between GCA and SATA. Reservations may also be
expressed regarding sampling of the biopsies. "Skip"
areas are recognized in GCA (27), and there is presently no evidence extant to indicate that SATA is
BRIEF REPORTS
entirely free from such uneven distribution. However,
the multiple blocking a n d cutting levels of the specimens should have minimized errors from this source.
In spite of these reservations, the data on GCA
closely match the findings in previously published reports in which corrections were made for the base
populations. Direct comparisons with SATA in the
same base population therefore seem valid. It may thus
be concluded that the patterns of occurrence and progression in GCA are very different from those of SATA,
and that there is no fundamental connection between
the 2 entities, other than that afforded by the coincidence of tissue location a n d old age.
1189
11.
12.
13.
14.
15.
Acknowledgments. I am indebted to Professor
Charles Goldsmith, Department of Epidemiology and Biostatistics, McMaster University, for advice and criticism,
and for the assistance of H. DiFrancesco in the preparation
of the manuscript.
16.
17.
REFERENCES
1. Meretoja J, Tarkkanen A: Amyloid deposits of internal
2.
3.
4.
5.
6.
7.
8.
9.
10.
elastic lamina in temporal arteritis. Ophthalmologica
170:337-344, 1975
Reid AH, Maloney AFJ: Giant cell arteritis and arteriolitis associated with amyloid angiopathy in an elderly
Mongol. Acta Neuropathol (Berl) 27: 131-137, 1974
Probst A, Ulrich J: Amyloid angiopathy combined with
granulomatous angiitis of the central nervous system:
report on two patients. Clin Neuropathol 4:250-259,
1985
Shintaku M, Osawa K, Toki J, Maeda R, Nishiyama T:
A case of granulomatous angiitis of the central nervous
system associated with amyloid angiopathy. Acta Neuropathol (Berl) 70:340-342, 1986
Gilmour JR: Giant-cell chronic arteritis. J Pathol Bacterial 53:263-277, 1941
Heptinstall RH, Porter KA, Barkley H: Giant-cell (temporal) arteritis. J Pathol Bacteriol 67:507-519, 1954
Dunea G, Owens FJ, Mackenzie AH: Polymyalgia rheumatica and renal amyloidosis: report of a case. Cleve
Clin Q 35: 193-197, 1968
Ostberg G: Morphological changes in the large arteries
in polymyalgia arterica. Acta Med Scand (suppl 533)
192:133-164, 1972
Goodman BW Jr: Temporal arteritis. Am J Med 67:839852, 1979
Boesen P, Sorensen SF: Giant cell arteritis, temporal
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
arteritis, and polymyalgia rheurnatica in a Danish
county: a prospective investigation, 1982-1985. Arthritis
Rheum 30:294-299, 1987
Muckle TJ: Senile amyloidosis of arterial elastica interna, Protides of the Biological Fluids. Edited by H
Peeters. Brugge, Oxford Pergamon Press, 1973
Puchtler H. Sweat F, Levine M: On the binding of
Congo red by amyloid. J Histochem Cytochem 10:355364, 1962
Wright JR, Calkins E, Humphrey RL: Potassium permanganate reaction in amyloidosis: a histologic method
to assist in differentiating forms of this disease. Lab
Invest 36:274-281, 1977
Llewellyn BD: An improved Sirius red method for
arnyloid. J Lab Med Techno1 27:308-309, 1970
Leung P, Muckle TJ: Progressive combined amyloidelastic stain. Arch Pathol 99:618-620, 1975
Wolman M: Amyloid, its nature and molecular structure: comparison of a new Toluidine blue polarized light
method with traditional procedures. Lab Invest 25: 104110, 1971
Adams CW: A para-dimethylaminobenzaldehyde-nitrite
method for the histochemical demonstration of tryptophane and related compounds. J Clin Pathol 10:56-62,
1957
Tukey JW: Exploratory Data Analysis. Reading, MA,
Addison-Wesle y , 1977
Mosteller F, Tukey JW: Data Analysis and Regression:
A Second Course In Statistics. Reading, MA, AddisonWesley, 1977
Fleiss JL: Statistical Methods for Rates and Proportions. Second edition. New York, John Wiley & Sons,
1981
Downie NM, Heath RW: Reliability, validity, and item
analysis, Basic Statistical Methods. Fourth edition. New
York, Harper & Row, 1974
Yates F: Contingency tables involving small numbers
and Chi2-test. J R Stat SOC(suppl) 1:217-235, 1934
Rosner D: Fundamentals of Biostatistics. Boston, Duxbury Press, 1982
Iman RL: Graphs for use with the Lilliefors test for
normal and exponential distributions. Am Stat 36: 109112, 1982
Downie NM, Heath RW: Other correlational techniques, Basic Statistical Methods. Fourth edition. New
York, Harper & Row, 1974
Wonnacott TH, Wonnacott RJ: Multiple regression extensions, Regression: A Second Course in Statistics.
New York, John Wiley & Sons, 1981
Klein RG, Campbell RJ, Hunder GG, Carney JA: Skip
lesions in temporal arteritis. Mayo Clin Proc 5 1 504-5 10,
1976
Документ
Категория
Без категории
Просмотров
2
Размер файла
381 Кб
Теги
elastica, inflammation, laminar, giants, arteries, amyloidosis, temporal, compare, interna, cells
1/--страниц
Пожаловаться на содержимое документа