To the Editor.—We read with great interest the recently published article by Strauchen and Miller.1 This paper pertains to the diagnostic use of immunohistochemical stains in infarcted lymph nodes and confirms many of our results, recently presented in an abstract form.2 This work and other reports clearly document that immunohistochemical stains are useful in elucidating the underlying processes in infarcted lymph nodes.1–5 

Drs Strauchen and Miller included 2 cases with a benign pattern of staining and a clinical follow-up negative for lymphoma.1 It may be construed that this pattern signifies a benign condition. However, one of the infarcted lymph nodes in our series had a similar staining pattern suggestive of normal lymphoid compartments while a partial involvement of this and the remaining viable lymph nodes by a blastic variant of mantle cell lymphoma was present.2 Therefore, the staining pattern in infarcted lymph nodes can lack morphologic details and can be deceiving in partial node involvement.

A difference of opinion can arise when the performance of different antibodies in necrotic tissue is compared. For example, Strauchen and Miller1 observed that the preservation of CD79a was superior to that of CD20 in infarcted tissue. Our experience is different. In our series, CD20 staining results had excellent agreement with the viable tissue results. However, CD79a usually failed to stain necrotic tissue.2 We are not sure how to explain this discrepancy, especially since we also used Dako (Dako Corporation, Carpinteria, Calif) CD20 (L26) and CD79a (HM-57) monoclonal antibodies. It is possible that differences in the antigen retrieval and staining protocol may account for this disparity. Some, but not all, of the previously published papers also documented a rather poor performance of CD20 in necrotic tissue.3–5 But when we pursued a human tonsil autolysis study, CD79a was detectable only up to 24 hours of autolysis, whereas CD20 was still detectable at a low intensity at 120 hours (maximum studied interval).2 

Our study indicates that occasional false-positive and false-negative results can be expected in necrotic lymphoid tissue.2 Thus, we conclude that the interpretation of immunohistochemical stains in infarcted lymph nodes is possible but that there are some limitations. The difficulties are compounded by the fact that the performance of some immunohistochemical stains may vary with different methods.

Strauchen
,
J. A.
and
L. K.
Miller
.
Lymph node infarction.
Arch Pathol Lab Med
2003
.
127
:
60
63
.
Polski
,
J. M.
,
C. H.
Dunphy
,
H. L.
Evans
, and
D. S.
Brink
.
Antigen detection in necrotic lymphoid tissue [abstract].
Mod Pathol
2001
.
14
:
176A
.
Pelstring
,
R. J.
,
D. C.
Allred
,
R. J.
Esther
,
S. R.
Lampkin
, and
P. M.
Banks
.
Differential antigen preservation during tissue autolysis.
Hum Pathol
1991
.
22
:
237
241
.
Norton
,
A. J.
,
A. D.
Ramsay
, and
P. G.
Isaacson
.
Antigen preservation in infarcted lymphoid tissue.
Am J Surg Pathol
1988
.
12
:
759
767
.
Kojima
,
M.
,
S.
Nakamura
,
S.
Sugihara
,
N.
Sakata
, and
N.
Masawa
.
Lymph node infarction associated with infectious mononucleosis: report of a case resembling lymph node infarction associated with malignant lymphoma.
Int J Surg Pathol
2002
.
10
:
223
226
.