When suspected tumors are directly visualized at fiberoptic bronchoscopy, we routinely obtain cytologic specimens with a bronchial brush before performing biopsies. Typically, these brush specimens are smeared onto a microscope slide premoistened with fixative and the brush discarded. For the purposes of this study, before the brush was discarded, it was agitated in an SBW vial containing fixative. This material, which would have otherwise been discarded with the brush, was then analyzed for content of cells and quality of cellular preservation. Results were compared with the results of the matched slides prepared using the direct smear technique (DST).
Twelve patients with apparent gross tumor visualized at broncho-scopic study underwent routine bronchial brushing with a disposable cytology brush (Microvasive No. 1BT/1.73/58/140). In each case the sheathed brush was passed via the suction port of the fiberoptic bronchoscope to a position adjacent to the lesion. The brush was extended and the lesion vigorously abraded in a back and-loi th manner exactly ten times. The brush was pulled back into the sheath, which was then removed from the bronchoscope. The brush was immediately extended out of the sheath and two direct smears made by firmly rolling the brush on microscope slides that had been premoistened with 95 percent alcohol. Each smear was then immediately immersed in the same fixative.
On completion of this process and just before discarding the brush, it was agitated in an up-and-down fashion in an SBW vial filled with a polyethylene glycol (Carbowax) solution (50 perccnt alcohol with 2 percent Carbowax). Subsequently, this vial was placed in a centrifuge and spun at 1,500 rpm for 12 minutes. All but two drops of the supernatant fixative was removed with a micropipet. The cell pellet was resuspended in the remaining fixative by gentle agitation. This suspension was then applied to two microscope slides. These smears were allowed to air dry completely and then immersed in 95 percent alcohol for at least 15 minutes. In an effort to avoid sampling error, slides used for both smear techniques had equal-sized, premeasured areas lor smearing designated by a wax pencil marking. buy levlen online
Both sets of smears (the DST and the SBW) were stained by a modified Papanicolaou staining procedure, coverslipped, and screened. The smears were reviewed by a eytotechnologist and a pathologist for the presence of diagnostic material. The results of these cytologic evaluations were compared with the results of bronchial biopsy and with any other diagnostic material.
Following confirmation of the presence of malignant cells, one of the SBW slides and one of the DST slides w;is each randomly selected for quantitative evaluation. The area of the smear on each slide was divided into ten sectors, each measuring 4 X 10 mm. These sectors were of equal size on all slides used in this study. An area from the right side, center, and left side of each slide (30 percent of the entire slide) was counted for all malignant cells using the high, dry (40 X) objective lens. From these counts the total number of cells per slide was extrapolated. In a few cases where malignant cells were quite scanty, the entire slide was counted to avoid sampling error.
Finally, paired SBW’-DST slides were reviewed in a blinded fashion by a cytotechnologist and a pathologist who graded each slide lor quality of preservation and cellular detail.
Because of the abnormal distribution of the results, a nonpara-metric method was used in the analysis ol the quantitative data. The Wilcoxon rank sum test was used, with statistical significance being defined as the equivalent of p<0,02.