The focus of "atypical glands, suspicious for malignancy" in prostatic needle biopsy specimens: incidence, histologic features, and clinical follow-up of cases diagnosed in a community practice.

This study was undertaken to determine retrospectively the prevalence and histologic features of the atypical foci that are suspicious for but are not diagnostic of a malignancy in contemporary prostate needle biopsy specimens reported in a community practice. Histologic features were examined in detail to identify features that prevented an unequivocal diagnosis of carcinoma. Of 1,009 prostate needle biopsy specimens obtained between January 1, 1993, and August 1, 1995, the diagnosis of an atypical focus suspicious for malignancy was made in 48 (4.8%). In review of the biopsy specimens diagnosed as benign, an additional 7 cases (0.7%) were identified. The following histologic features were identified in 54 cases: enlarged nucleoli, 54 (100%); enlarged nuclei, 45 (83%); intraluminal eosinophilic secretions, 40 (74%); infiltrative growth, 37 (68%); small acinar proliferation, 37 (68%); intraluminal basophilic mucin, 23 (42%); amphophilic cytoplasm, 18 (33%); high-grade prostatic intraepithelial neoplasia, 17 (31%); and crystalloids, 12 (22%). Corpora amylacea were not identified. The foci contained from 1 to 67 acini (mean, 20.7). Although each atypical focus showed most of the features of adenocarcinoma, an unequivocal diagnosis of malignancy was not given owing to four features: the small size of the focus, the small number of cells with enlarged nucleoli, the clustered growth pattern, and the presence of high-grade prostatic intraepithelial neoplasia within many of the foci. At initial examination, 36 of 41 patients (83%) had an elevated serum concentration of prostate-specific antigen (mean, 10 ng/mL), and 20 (49%) had abnormal findings on a digital rectal examination. Twenty-five patients (46%) underwent additional sampling of the prostate, and 15 of these (60%) were found to have adenocarcinoma; the remaining 30 patients did not undergo a subsequent biopsy. Patients with subsequent cancer had higher mean serum concentrations of prostate-specific antigen and change in concentrations of prostate-specific antigen than those whose repeat biopsy results were negative; no other clinical or histologic differences were observed between these two groups. To the community pathologists in this study, the lack of prominent nucleoli, the small size of the focus, clustered acini, and/or adjacent high-grade prostatic intraepithelial neoplasia prevented an unequivocal diagnosis of malignancy. If a prostate needle biopsy specimen is reported as containing an atypical focus suspicious for malignancy, a subsequent biopsy is warranted given the high predictive value for adenocarcinoma.

A hemolytic reaction implicating Sda antibody missed by immediate spin crossmatch.

Anti-Sda, an antibody not usually considered to cause of hemolytic transfusion reactions, possibly was related to hemolysis following transfusion of red blood cells expressing strong Sda antigen. Prior to transfusion, the antiglobulin antibody screen performed in LISS and an immediate spin crossmatch were negative. Retrospectively, after hemolysis was detected, an antiglobulin crossmatch with red cells from the transfused unit revealed microscopic incompatibility. The transfused unit proved to have strong expression of Sda antigen-facilitating identification of a weak Sda antibody in our patient. In addition, this case represents an unusual instance in which an antibody screen plus an immediate spin crossmatch failed to detect an incompatibility that would have been apparent had an antiglobulin crossmatch been performed.

Cytomegalovirus detection in bone marrow transplant patients with idiopathic pneumonitis. A clinicopathologic study of the clinical utility of the polymerase chain reaction on open lung biopsy specimen tissue.

The morbidity and mortality rates of cytomegalovirus (CMV) infections, including pneumonitis in bone marrow transplant patients, are well documented and yet no rapid, sensitive diagnostic tool is available. To test the polymerase chain reaction as such a diagnostic tool, formalin-fixed, paraffin-embedded open lung biopsy material was selected from post-bone marrow transplant patients in whom pulmonary parenchymal changes were evident but viral inclusions were not seen. As a control, other immunosuppressed (non-bone marrow transplant) patients and low-risk nonimmunosuppressed patients were studied in a similar manner. Viral culture results and clinical data were available on all the high-risk patients. Two of 15 high-risk patients were found to have amplifiable CMV DNA despite the lack of histologic viral inclusions. One of these patients had been treated recently for CMV pneumonia, and the other had a positive culture of the lung specimens for CMV 13 days after biopsy. None of 12 control patients had amplifiable CMV DNA. These data indicate that the polymerase chain reaction is more sensitive than histologic examination and at least as sensitive as viral culture without evidence of false-positive results.

Cytologic criteria used to diagnose adenocarcinoma in pleural effusions.

A stepwise logistic regression analysis was employed to evaluate the usefulness of the following criteria to distinguish adenocarcinoma in pleural effusion from benign pleural effusion: increased size of nucleus, increased nuclear to cytoplasmic ratio, irregular nuclear borders, sharply defined cytoplasmic boundaries, large nucleoli, aggregates with variable size nuclei, aggregates with nuclear overlap, irregular noncentral vacuoles, aggregates with nuclear molding, multinucleation, three-dimensional aggregates, aggregates with large cytoplasmic vacuoles, atypical mitoses, nuclear vacuoles, homogeneous cytoplasm, aggregates with associated lymphocytes and neutrophils, and uniform size aggregates. A total of 223 patients with benign pleural effusion cases and 221 patients with adenocarcinoma in their pleural effusion were scored as to the presence or absence of the above criteria. The resulting data were subjected to a stepwise logistic regression analysis that chose increased nuclear/cytoplasmic ratio, irregular nuclear borders, large nucleoli, sharply defined cytoplasmic boundaries, and three-dimensional aggregates as the best criteria to differentiate adenocarcinoma in pleural effusion from benign pleural effusion. When used together, these five features had a sensitivity of 94% and a specificity of 93% for predicting adenocarcinoma.

Renal biopsy frozen section: a fluorescent study of hematoxylin and eosin-stained sections.

When frozen sections of kidney are stained with hematoxylin and eosin (H&E) and viewed by standard fluorescence microscopy, crisp detail of nephron and vascular basement membranes is revealed. We studied by H&E fluorescence, control kidneys, and 85 renal biopsies which span a broad range of renal disease. We found that H&E fluorescence provides useful initial orientation by revealing the presence of cortex or medulla and number of glomeruli and vessels. It also reveals major structural alterations of nephron and vascular basement membranes (sclerosis, disruption, atrophy) and heavy deposition of abnormal material (fibrin, immune deposits, paraprotein). When coupled with direct immunofluorescence, a much greater understanding of the disease process is possible, prior to review of permanent sections.

Anionization of an antigen promotes glomerular binding and immune complex formation.

Bovine serum albumin (BSA, pI 4.9) was maleylated to yield highly anionic MBSA (pI 3.0). Maleylation of BSA lead to an expansion of molecular size of native BSA from an effective molecular radius (EMR) of 37 A to 57 A for MBSA as assessed by gel filtration chromatography. MBSA, but not BSA, bound to the peripheral capillary wall (PCW) and mesangium in vitro in frozen sections, and in vivo following i.v. injection (0.006 mg/g body wt), examined by immunofluorescence. When similarly injected rats or controls were given antibodies to either MBSA or BSA following injection of antigen, immune complexes were observed in glomeruli by immunofluorescence and EM only in MBSA injected rats. Deposits occurred in the mesangium and subendothelium in the PCW. In frozen sections, bound MBSA could be partially removed from tissue sections by high ionic strength buffer. Also, binding of MBSA was diminished by prior treatment of sections with synthetic polyanions. Maleylated bovine gamma-globulin and succinylated BSA showed identical binding patterns as described for MBSA, indicating that binding was not unique to the modified BSA molecule nor to the form of anionization. These results indicate that charge interactions between circulating highly anionic macromolecules and cationic domains within glomerular structures are responsible, in part, for MBSA binding and subsequent localization of immune complexes. Furthermore, it is inferred that the selective binding of MBSA to glomeruli and formation of immune complexes occurred by a mechanism not related to difference in size between MBSA and BSA. These findings are different from conventionally understood charge interactions in glomerular immune complex formation.