An official research policy statement of the American Thoracic Society/European Respiratory Society: Standards for quantitative assessment of lung structure

Connie C W Hsia, Dallas M. Hyde, Matthias Ochs, Ewald R. Weibel

Research output: Contribution to journalReview articlepeer-review

645 Scopus citations


The charge of this Joint ATS/ERS Task Force was to critically review the state-of-the-art stereological methods in lung morphometry, provide practical guidelines for use of these methods in basic and translational lung research, define standards to promote comparability of morphometric studies, and examine the extension of these methods to noninvasive lung imaging. Broad conclusions regarding study design and standardization: 1. In quantitative assessment of lung structure, accuracy is far more critical than precision, because inaccurate or biased data cannot be made accurate by increasing the number of measurements. The only effective way to avoid bias and ensure accuracy is via rigorous experimental design and standardization of each step of tissue fixation, processing, sampling, and analysis. 2. With an efficient study design, the number of samples, sections, images, and measurements at each analytical stage can be kept low without compromising accuracy and still achieving reasonable global precision in the results. 3. Principles of design-based stereology can be applied to the sampling and morphometric analysis of structures obtained by in vivo imaging modalities such as computed tomography (CT), magnetic resonance imaging (MRI), and positron emission tomography (PET). Principles for standardization of study design: 4. The lung should be fixed under well-defined inflation and perfusion conditions, using appropriate fixative(s) and processing procedures that result in the best structural preservation with the least tissue distortion for the intended study goal. 5. Tissue blocks or image fields should be selected using proven unbiased sampling schemes to ensure that all parts of the whole have an equal chance for being sampled. 6. Structures that are nonrandom in orientation or distribution, and biopsy samples from nonrandom target sites, require special sampling and analytical strategies. Recommendations for methods of quantification: 7. Stereological methods that are free of geometric assumptions must be used to efficiently quantify number, length, surface area, and volume at an adequate spatial resolution to ensure that measurements made on two-dimensional (2D) images accurately represent the three-dimensional (3D) structure. 8. Alveolar surface area can be accurately estimated from profile boundary lengths or intersection counting using systematic linear probes in 2D sections that are "uniform random" (i.e., selected from all possible sections with the same probability); however, the measurement is sensitive to resolution. 9. Alveolar number and size (volume) can be accurately measured using 3D volume probes such as the disector, but not from simple counts of profiles or measurements of cross-sectional areas in uniform random 2D sections. 10. Stereological measurements should be related to the volume of the lung or an appropriate reference space. Measurements made on lung biopsy specimens should be related to an internal reference space.

Original languageEnglish (US)
Pages (from-to)394-418
Number of pages25
JournalAmerican journal of respiratory and critical care medicine
Issue number4
StatePublished - Feb 15 2010


  • Cell number
  • Cell size
  • Cell volume
  • In vivo imaging
  • Morphometry
  • Reference lung volume
  • Stereology
  • Surface area
  • Unbiased sampling

ASJC Scopus subject areas

  • Pulmonary and Respiratory Medicine
  • Critical Care and Intensive Care Medicine


Dive into the research topics of 'An official research policy statement of the American Thoracic Society/European Respiratory Society: Standards for quantitative assessment of lung structure'. Together they form a unique fingerprint.

Cite this