mgh:cyto-1-4

Body Cavity Fluids Cases:

Ivan Chebib MD, Amy Ly MD, Ron Arpin SCT

• Indications for cytology examination

by Amy Ly, M.D.
The pleural, pericardial, and peritoneal cavities are lined by serosa, which is a simple layer of mesothelial cells. Under normal conditions, these cavities contain only a small amount of fluid which allows adjacent serosal surfaces to move over each other with low resistance during normal organ activities (e.g. breathing, heartbeats, peristalsis). In disease states, a greater amount of fluid accumulates and is called an effusion. Effusions may be characterized clinically as transudative or exudative. Transudates result from unbalanced hydrostatic and oncotic pressures. Exudates result from injury to the mesothelium, which is commonly caused by malignant tumors that have spread to serosal surfaces and/or malignant mesothelioma that originates in the serosa.

Detection of serosal malignancy by cytologic exam is more sensitive than by blind biopsy (58%-71% compared with 45%). Cytology sensitivity is further increased by 2%-38% if more than one sample is examined. However, the false negative rate is still significant. If cytology is negative but there is high suspicion for pleural malignancy, the patient can undergo thoracoscopy for further evaluation.

The specificity of cytologic effusion evaluation is very high: the false positive rate is <1%. False positive and false suspicious diagnoses are mainly due to reactive mesothelial cells that appear atypical.

Gynecologic and non-gynecologic malignancies involving the peritoneal serosal surfaces may not produce an effusion or be associated with lesions visible by gross inspection intraoperatively. In such cases, the peritoneal cavity may be evaluated by “peritoneal washing,” which is part of a cancer staging procedure. Peritoneal washings may also be used to exclude occult malignancy in patients undergoing laparoscopy or laparotomy for presumed benign gynecologic conditions and in women with BRCA1/2 mutations undergoing risk reducing salpingo-oophorectomy. Peritoneal washing may be potentially utilized to monitor a patient's response to adjuvant treatment for cancer.

Peritoneal washings that are positive for malignancy are associated with worse prognosis in patients with ovarian and fallopian tube cancers. Positive washings alone impact the surgical stage of only 3-5% of women with gynecologic cancers, but may be the only evidence of metastasis to the peritoneum for some patients. Peritoneal washing results are included in the International Federation of Gynecology and Obstetrics ovarian and fallopian tube cancer staging algorithm. The prognostic utility of this test for endometrial and other gynecologic cancers is unclear at this time.

There is a significant false-negative rate with peritoneal washings. 23-86% of patients with biopsy proven peritoneal metastasis have no evidence of disease in their washings by cytologic exam. The high false negative rate may be partly due to poor distribution of fluid within peritoneal cavities that have been affected by adhesions. False positive diagnoses are not common (<5% of cases), and are usually due to proliferative mesothelial cells with reactive changes and associated psammoma bodies, and endometriosis.

• Procuring the specimen

by Amy Ly, M.D.
Effusion specimens are obtained by inserting a needle into the pleural space (thoracentesis), pericardial space (pericardiocentesis), and peritoneal cavity (abdominal paracentesis). Peritoneal fluid is usually obtained through the abdominal wall, however in women it may also be aspirated from the cul-de-sac through the vagina (cold to centesis). Effusions may also be collected during thoracic, abdominal, or cardiac surgery. Removing this excess fluid may be performed for diagnostic purposes (submitted for pathology evaluation) or therapeutic purposes (to alleviate symptoms such as dyspnea and heart failure). Large volumes (several liters or more) of abdominal fluid may be drained safely. However, pleural fluid that is rapidly removed in large quantities may rarely be complicated by reexpansion pulmonary edema. This condition is fatal in up to 20% of cases and tends to involve younger patients with a long duration of lung collapse who experience rapid lung reexpansion upon thoracentesis.

The effusion is collected in sterile containers and sent unfixed to the laboratory. Specimen collection into glass containers causes rapid clotting, which is undesirable as this causes dispersion of cells and makes it more difficult to evaluate them. To prevent clotting, collect fluids into heparinized bottles containing 3 units of heparin per milliliter of capacity. If heparinized bottles are not available, the heparin should be placed into the container before the fluid is drained. Store fluids at 4°C until the time of slide preparation. Effusions are robust specimens and may be refrigerated for > 2 weeks without compromising cellular morphology or antigenicity for immunostains because the effusion itself nourishes the cells within it. However, specimens involved by malignancies with high cellular turnover (e.g. Burkitt lymphoma) should be prepared as soon as possible.

Peritoneal washes are obtained intraoperatively. The surgeon evacuates any pre-existing peritoneal fluid and submits it separately for cytologic examination. Sterile saline (50-200 mL) is instilled into multiple areas, usually the pelvis, the right and left paracolic gutters, and the undersurface of the diaphragm. A repeat washing or rinsing action is used to abrade cells from the serosal surfaces into the saline. The saline is then pooled into a single collection and heparinized. There is no advantage to submitting washings from different sites separately. The specimen should be delivered to the laboratory unfixed and stored at 4°C until slides can be prepared. If there will be a significant delay before slide preparation, an equal volume of 50% ethanol can be added to preserve the specimen.

• Test platforms/specimen processing and triage

by Amy Ly, M.D.
To make slides from an effusion, the first steps are to agitate the fluid to evenly disperse the cells and then to centrifuge up to 50 mL of the fluid. The supernatant is discarded and the pellet is used to prepare smears, cytocentrifuge preparations (Cytospins), or thin-layer preparations (e.g. ThinPrep, SurePath). The slides are usually alcohol fixed but if a lymphoproliferative disorder is suspected, air dried Cytospins are helpful. Slides are stained with a Papanicolaou or Romanowsky type stain. Residual fluid is set aside in case additional slides or other preparations/tests such as cell block, flow cytometry, and molecular studies are needed.

Cell blocks may be prepared from fluids by coagulating the sediment into a compact mass with plasma and thrombin, wrapping the sediment in filter paper, placing in a cassette, and processed in the manner of histologic sections (fixing in formalin, embedding in paraffin, cutting, and staining with H&E). Clots that are already present in the fluid because it was not heparinized should be placed in cassettes for processing as cell blocks. The addition of a cell block to a smear/Cytospin/thin-layer slide increases sensitivity for the detection of malignancy. Cell block sections are useful for special and immunohistochemical stains. Cell block sections are also convenient for morphologic comparison with histologic sections because the tissues have been processed in an identical manner.

To prepare slides from a peritoneal washing, the specimen is thoroughly mixed and 50 mL of fluid is centrifuged. The supernatant is discarded and the pellet can be used to prepare smears, cytocentrifuge preparations (Cytospins), or thin-layer preparations (e.g. ThinPrep, SurePath). The remaining material or a separately centrifuged cell pellet can also be fixed in 10% formalin and processed as a cell block, employing histologic methods of processing, paraffin embedding, cutting, and H&E staining. Cell block sections are useful for morphologic comparison to the patient's resected neoplasm and for performing special and immunohistochemical stains.

• Reporting and terminology

by Amy Ly, M.D.
There are no established criteria for adequacy of effusion specimens. Cytologic diagnosis of fluids utilizes the following categories: “no malignant cells identified,” “atypical” (low suspicion for malignancy), “suspicious” (high suspicion for malignancy), and “positive for malignant cells.” The diagnosis of malignancy is semi-quantitative and semi-qualitative. “No malignant cells identified” and “positive for malignancy cells” are self-explanatory unequivocal diagnoses. Indeterminate categories of “atypical” and “suspicious for malignancy” are used when abnormal cells are present, but are too poorly preserved or too few in number to render a definitive diagnosis of malignancy. Approximately 5% of specimens are diagnosed as “suspicious.” In such cases, the effusion will usually re-accumulate if there is a serosal malignancy; the next sample may contain evidence of malignancy.

Adequacy criteria for peritoneal washing cytologic specimens have not been established, but there should be at least a few groups of well-preserved benign mesothelial cells present before concluding that the specimen is adequate for evaluation and negative for malignant cells. Specimens with malignant cells are always adequate. Results of peritoneal washing cytology are commonly reported as negative, atypical, suspicious, or positive for malignant cells. Atypical and suspicious interpretations should be avoided if possible because they are not helpful for treatment decision-making. Usually, only an unequivocally positive diagnosis is used for staging purposes, and atypical and suspicious results are considered to be negative results. Equivocal cytology washing cases may be resolved by comparing morphology a current corresponding resection specimen.

Basic cytomorphology:

Normal mesothelial cells – MN05-G13557

• Sheets of benign mesothelial cells are often smaller than 12 cells, but may sometimes be composed of upwards of 50 cells
• In these photomicrographs, the even dispersal of uniform cells, with regular nuclei, delicate nuclear membranes and small round nucleoli signal the benign nature of these cells
  
  

Reactive mesothelial cells – N13-8012

• Under conditions of an inflammatory process, mesothelial cells are increased in number, can exhibit a wide range of sizes, and may be multinucleated
• The keys to diagnosis involve (1) applying individual criteria of benignity and (2) establishing the presence of an uninterrupted continuum of sizes from small to very large
• Note enlarged nuclei, small multiple nucleoli, and spaces between adjacent cells, so called "windows"
• Inflammatory cells are present in the background
• Like pleural effusion, mesothelial cells in peritoneal effusions may exhibit a range of cell sizes
• Mesothelial cells may be admixed with inflammatory cells and histiocytes.
  
  

Mesothelioma – N12-12597

• The key to diagnosing mesothelioma is not identifying a second malignant cell population
• Final determination may require immunocytochemistry or a cell block with immunohistochemistry, electron microscopy, or other specialized techniques
• Individual malignant mesothelial cells exhibit a rim of ruffled, less dense cytoplasm (ectoplasm), surrounding dense cytoplasm around the nucleus (endoplasm)
• Tumor cells may be seen in a background of blood and proteinaceous debris
• Groups of more than 12 cells may be a feature of malignancy.
• High N/C ratio with variability in nuclear size and occasional multi-nucleation confirm the malignant nature of these cells
• Differential diagnoses include adenocarcinoma and mesothelioma
• Fine microscopic features of peripheral cell membranes and intercellular windows may suggest mesothelioma
• Abnormal mitotic figures may be noted with mesothelioma, other malignancies, as well as occasional reactive mesothelial cells in effusions
  
  

Metastatic adenocarcinoma – lung – N13-7980

• Papillary glandular arrangements of the tumor cells
• Prominent nucleoli, vacuolization and mitotic figures
• Distinctions from other sources of adenocarcinoma may be impossible.
  
  

Metastatic adenocarcinoma – breast – C99-T533

• Metastatic ductal carcinoma cells exhibits large irregular nuclei and nucleoli
• The classic description of metastatic breast cancer in pleural effusions employs the term "cannonballs" to emphasize the rounded arrangement of tumor cells
• They may have a relatively small nuclear size
• Nuclei are vesicular with prominent nucleoli
• Cytoplasmic vacuoles are uncommon
• A cell block of the cells allows for assay of hormonal receptors or other epithelial markers, such as her-2-neu
  
  

Metastatic adenocarcinoma – ovary – N13-6042 and N13-5843

• Cells of papillary serous ovarian adenocarcinoma in a pleural effusion represent a discontinuous population of cells
• Their cell and nuclear size is variable
• Increased nuclear to cytoplasmic ratio and cytoplasmic vacuoles are features
• Cells may exist singly or in small acinar groups
• Vigorous peritoneal washes may dislodge microscopic tumor
• Washes are an integral part of staging laparoscopy
• Because of the washing procedure, tumor cells generally come off in three- dimensional cohesive groups and may be admixed with sheets of benign mesothelium
• The tumor cells are easily distinguished by size, malignant characteristics and crowded configurations
  
  

Metastatic adenocarcinoma – GI –C98-N39001

• Gastric adenocarcinoma
• Cells with malignant features are present as a distinct population
• Some may exhibit nuclear displacement by a large secretory vacuole, a "signet ring" cell
• Origin from one part of the GI tract over another cannot be easily ascertained
• Cholangiocarcinoma, either from an intra-hepatic source or from an extra-hepatic biliary tree, may look like adenocarcinoma from elsewhere in the GI tract
• By exclusion of other sources through endoscopy, ultrasonography and/orCT imaging, the location may be determined.
  
  

Melanoma –C99-W27742

• Dyshesive single cells
• Malignant nuclear features, eccentric nuclei
• Range of patterns: small, spindle or epithelioid cells
• Nuclear size variation
• Nuclear pseudoinclusions with bi-, and multinucleation
• Intracytoplasmic dusty brown melanin pigment
• S-100, HMB-45, Melan-A positive (not always)
  
  

Lymphoma – N13-6082

• Dyshesive single cells
• Open granular chromatin
• Nucleoli based on nuclear membrane in some subtypes
• Nuclear membrane protrusions and irregularity
• Scant cytoplasm in some subtypes (high N/C ratios)
• Lymphoglandular bodies in background
• LCA positive, B or T cell lineage