The aim of this study was to assess the validity of endoscopic bronchial biopsy specimens for the quantitation of nerves. To this end, endobronchial biopsy was simulated ex vivo on surgically resected lung specimens and nerve densities were compared in airway smooth muscle of biopsy and surrounding tissue, Specimens were stained immunohistochemically for the general neural marker protein gene product 9.5 (PGP 9.5) and for vasoactive intestinal peptide (VIP), and nerve densities were quantitated using computer-assisted image analysis, Nerve density for total (PGP 9.5-immunoreactive) nerves was slightly higher in biopsies than in corresponding lung tissue, but this difference did not reach statistical significance (p=0.08). There was also no significant difference in the density of VIP-immunoreactive nerves (p=0.60). These findings support the use of endobronchial biopsy specimens to quantitate nerves in asthma and other airway diseases. Copyright (C) 2000 John Wiley & Sons, Ltd.

A technique is proposed for applying well-established stereological methods to study fixed nasal biopsy material to obtain an unbiased estimate of blood vessel surface and volume densities. Biopsies of the nasal mucosa from the anterior 10 mm of an inferior turbinate were obtained from 18 subjects (15 males, 3 females with a mean age of 28.5 years [range: 17-54 years]), ten of whom had perennial allergic rhinitis, and eight control subjects. The mucosal tissue volumes were estimated by water displacement. Zamboni's-fixed cryostat sections (10 microns thick), stained with haematoxylin and eosin, were examined histologically. Computerised images of randomly selected tissue sections were analysed with point-counting and intercept-counting techniques to determine large blood vessel volume and surface densities, respectively. There were no significant differences between the volumes of tissue analysed from the control and rhinitic subjects (p = 0.35). The average volume density of the vessels was similar in the control group (6.17 +/- 1.41%) and the rhinitic group (7.8 +/- 5.59%; p = 0.38), but with a greater variability in the rhinitic group. Surface density estimations were 3.14 +/- 0.74 mm-1 in the control group and 3.10 +/- 1.41 mm-1 in the rhinitic group. Therefore, on average, the volume and surface densities of the cavernous blood vessels in rhinitis were unaltered and there was no evidence of vascular remodelling.

At present, the stereological assessment of histological sections is made possible by the use of manual counting techniques which estimate measurement parameters. These methods are tedious, time-consuming, and subject to operator error. This paper describes a UNIX-based computer program, OpenStereo, which was developed to facilitate the quantitative investigation of innervation and vascularity from histological sections. We designed OpenStereo to reduce operator error and increase the efficiency of stereological point counting for volume estimation and intercept counting for surface area analysis. The program was written in the C language for the Sun Workstation and uses the XView graphics user interface. Digital images, obtained by a variety of modalities, may be processed using stereological point counting, interceptions, planimetry, or thresholding techniques. The program displays selected images in a random fashion for analysis or processing and records the number of manually selected points or interceptions. Delineation of the reference space provides the computer with the data necessary to calculate volume or surface densities. The efficiency of OpenStereo was demonstrated by performing a pilot study on the quantification of innervation in the normal human colon and ileum. This stereological package benefits from the features of the X-windowing environment and has proved to be suitable for what has hitherto been a tedious and time-consuming task.