Skip to main content
. 2004 Jun 29;91(3):558–563. doi: 10.1038/sj.bjc.6601942

Table 4. Diagnostic sensitivities of cytology, FISH, and the combination of both (cyto&FISH) in the whole series of effusions and major tumour entities (with ⩾10 effusions per tumour entity available).

Tumour Effusion (n) Sensitivity cytology (%) Sensitivity FISH (%) P Sensitivity Cyto&FISH (%) P
All 358 44.4 53.9 0.0001 60.9 0.0001
Gynaecologica 147 54.4 63.9 0.018 72.1 0.0001
 Breast 104 50 63.1 0.031 71.2 0.0001
 Ovarian 29 65.5 72.4 0.727 82.75 0.063
 Endometrial 11 54.4 63.6 1 63.6 1
Gastrointestinalb 87 28.7 36.8 0.118 41.4 0.001
 Pancreatic 34 38.2 52.9 0.125 55.9 0.031
 Hepatoma 17 0 5.9 1 5.9 1
 Gastric 14 50 64.3 0.625 71.4 0.25
 Colorectal 11 18.2 18.2 1 27.3 1
Lung 71 54.9 71.8 0.008 76.1 0.0001
UPCc 17 35.3 47.1 0.687 58.8 0.125
Haematologicald 10 50 30 0.5 50 1

The P-values indicate significance levels for the comparison of FISH vs cytology (left column) and of cyto&FISH vs cytology (right column).

a

Gynaecologic tumours: breast (n=104), ovarian (29), endometrial (11), and cervical carcinoma (3).

b

Gastrointestinal tumours: pancreatic (34), hepatocellular (17), gastric (14), colorectal (11), cholangiocellular (9), and oesophageal carcinoma (2).

c

UPC, unknown primary cancer.

d

Haematological malignancies: non-Hodgkin's lymphoma (5), chronic myelogeneous leukaemia (2), Hodgkin's lymphoma (2), idiopathic myelofibrosis (1).