Conference on "Ground-Glass" Opacities

Saturday, September 9, 2000 to Sunday, September 10, 2000


Consensus Statement: 

The goal of this two-day conference, held at the National Cancer Center Hospital, was to develop a conceptual framework for the study of non-solid opacities detected in CT screening for lung cancer. This topic is important in that these non-solid opacities may represent precursor lesions of invasive adenocarcinoma. Participants of the conference included representatives from the National Cancer Center Hospital and the National Cancer Center-East Hospital who have participated in the assessment of CT screening for the Anti-Lung Cancer Association (ALCA) in Japan and from Early Lung Cancer Action Project (ELCAP) team at Weill Medical College of Cornell University in New York City, New York. Both groups had started CT screening for lung cancer in 1993 (1,2) and thus had long-term follow-up on the findings. They met to discuss the results of the past seven years of screening experience. The discussions focused on the radiologic appearance, diagnostic work-up and pathologic diagnosis of these non-solid opacities, with only preliminary considerations given to the choice of intervention.

It was the consensus of the group that these non-solid opacities should not be called "ground-glass opacities" (ggos) or "ground-glass attenuation," which are familiar terms in the radiologic literature. These terms are not properly descriptive and has been used with many diseases of the lung, diffuse as well as localized, such as pulmonary edema and diffuse pneumonia, among others. These opacities are actually transparent as they do not obscure the bronchial and vascular structures on CT images, in contrast to solid opacities. It is important to clearly refer to findings and determine which ones require further work-up in the context of CT screening for lung cancer. Key points in the discussion about the terminology was that it should be descriptive of the imaging findings and correlate with the corresponding pathologic findings.

Much discussion ensued, both at the meeting and after the meeting, in trying to find appropriate terminology which conveys the concept of a focal nodule which has a texture which can range from one extreme (e.g., opaque) to the other extreme (e.g., transparent, hazy, misty, lacy). An example of such terminology would be: opaque nodule, hazy nodule, and anything in between is a mixed nodule. As this is an important definition in the CT screening context, we decided to have further discussions on terminology at the upcoming workshop on this topic at the Third International Conference on Screening for Lung Cancer.

Both teams of investigators had identified two first-order subtypes of "ggos". The first consisted of "ggos" without any solid components, termed pure. The second consisted of "ggos: with solid components, termed mixed. Blood vessels within the "ggo", which may appear as solid components, are not to be considered in calling a pure "ggo" a mixed "ggo". Other potentially important descriptors of "ggos" include: 1) size (length, width, volume) of the entire "ggo" and and of its solid component; it was recognized that the solid component could be linear, rounded, or irregular; 2) presence/absence of spiculations or pleural tags; 3) other edge characteristics (sharp or blurred), with a sharp edge being subjectively defined as being able to trace out the edge with a pencil, and 4) the degree to which the "ggo" is opacified, although a meaningful metric was not defined. It was recognized that these descriptors also depend on the CT scanning and viewing parameters and that these must be clearly specified. It was also recognized that image-analytic techniques might in the future allow us to obtain more consistent documentation of these ggo descriptors.

These "ggos" were, in the context of CT screening, recognized by those who were performing screening as being potentially malignant. Not much is at yet known about their frequency of occurrence and frequency of malignancy, but considerable data have already accumulated. Preliminary review based on the two screening programs suggests that the malignancy rate of the mixed "ggos" may well be as high as those of solid nodules while that of the pggos appears to be much lower. For this reason, it was agreed that the recommended diagnostic work-up of screen-detected mixed "ggos" should be the same as that for solid nodules, beginning with high-resolution CT images (in sections of 1 mm or less) of the opacity to better characterize the mixed "ggo".

It was agreed that individuals with pure "ggos" of size (average of length and width) 5 mm or less should continue with scheduled screening without further diagnostic work-up until either growth or development of a solid component could be demonstrated. Individuals with pure "ggos" larger than 5 mm should have high-resolution CT imaging to determine whether solid components could be identified. If no solid components can be identified, further diagnostic work-up should be done on a case-by-case basis with either follow-up CT to determine any change in the size or density of the opacity or by biopsy.

Pathologic review of the biopsied and resected "ggos" all resulted in the diagnosis of either fibrosis (scar), atypical adenomatous hyperplasia (AAH), bronchioalveolar carcinoma (BAC) with or without pathologic features of collapse (3,4) or invasive adenocarcinoma. Although the preliminary data on the occurrence of AAH in the resected lobes of these malignancies suggest that AAH may be a precursor to BAC and that BAC may well progress to invasive adenocarcinoma, further research is needed to develop a better understanding of the disease progression. It appears that BAC without hilar or mediastinal lymph node involvement may have a higher cure rate (around 100%) than invasive adenocarcinoma (3-5), and thus long-term follow-up of those choosing early surgical intervention as compared with those choosing strict follow-up CT imaging may indicate the aggressiveness and curability of each type of malignancy. Critical to this research is careful radiologic-pathologic correlation including high-resolution imaging of the "ggos" prior to biopsy or surgical intervention, specimen radiography, determination of the pathologic features of the edge as well as of the non-solid and solid components when present, special stains of the specimen to determine disruption of the basement membrane, and in the future, assessment of pertinent biomarkers as these become better defined.

Recommended surgical intervention for solid nodules and mixed "ggos" continues to be lobectomy with complete mediastinal lymph node dissection. However, much discussion centered on the appropriate surgical intervention for pure "ggos", particularly those of size 6-20 mm. It was felt that for these pure "ggos", a small incision thoractomy with partial resection or segmentectomy may be sufficient and this procedure is being currently being assessed at the National Cancer Center-East Hospital. An elastin stain on frozen tissue can aid in the assessment of the intactness of the basement membrane to differentiate non-invasive BAC (Noguchi Type A or B) from BAC with microinvasion (Noguchi Type C) and, in Japan, this is being done prior to the decision to perform a lobectomy or a more limited resection (6,7).

It was agreed that further discussion of the terminology, criteria and long-term results is needed for "ggos". Critical to this assessment is the reproducibility of the radiologic classification and further efforts are being made by both groups on the development of teaching files. Workshops on these aspects will also be held at the upcoming conference in New York City at Weill Cornell Medical College of Cornell University on October 27-29, 2000 (8).

Japan United States of America
Hisao Asamura, MD Claudia Henschke, PhD, MD, Chair
Kenji Eguchi, MD Marek Kimmel, PhD
Ryutaro Kakinuma, MD Anthony Reeves, PhD
Masao Nakata, MD William Travis, MD
Mashihiro Kaneko, MD Madeline Vazquez, MD
Tomotaka Sobue, MD David Yankelevitz, MD
Kenji Suzuki, MD  
Junji Yoshida, MD  


Wentao Fang, Thoracic Surgery, Shanghai Chest Hospital, China

Joo Hyun Kim, Thoracic Surgery, Seoul National University, Seoul, Korea

Barbara Kimmel, Baylor Medical College, Houston, Texas

Henri Porte, Thoracic Surgery, Lille University Hospital, Lille, France

Daniel Vlock, Ethicon Endo-Surgery, Inc, US


  1. Kaneko M, Eguchi K, Ohmatsu H, Kakinuma R, Naruke T, Suemasu K, Moriyama N. Peripheral lung cancer: screening and detection with low-dose spiral CT versus radiography. Radiology 1996;201:789-802.
  2. Henschke CI, McCauley DI, Yankelevitz DF, Naidich DP, McGuinness G, Miettinen OS, Libby DM, Pasmantier MW, Koizumi J, Altorki NK, Smith JP. Early Lung Cancer Action Project: overall design and findings from baseline screening. Lancet 1999;354:99-105.
  3. Noguchi M, Moricawa A, Kawasaki M et al. Small adenocarcinoma of the lung. Histologic characterisation and prognosis. Cancer 1995;75:2844-2852.
  4. WHO Panel on Lung Cancer, Chair:William Travis, Armed Forces Institute of Pathology, Washington DC:
  5. Special Session of the WHO Panel on Lung Cancer Meeting, Chair:William Travis, Ninth World Conference on Lung Cancer, International Association for the Study of Lung Cancer (IASLC) and the Japan Lung Cancer Society. Tokyo, Japan, September 11, 2000.
  6. Yoshida J. Surgical resection of non-invasive bronchioalveolar carcinoma at the National Cancer Center-East Hospital. Presentation at the Conference on "Ground-glass Opacities," National Cancer Center, Tokyo, September 9, 2000.
  7. Suzuki K, Yokose T, Yoshida J, Nishimura M, Takahashi K, Nagai K, Nishiwaki Y. Prognostic Significance of the Size of Central Fibrosis in Peripheral adenocarcinomas of the lung. Ann Thorac Surg 2000;69:893-897.
  8. Third International Conference on Screening for Lung Cancer. October 27-29, 2000, Weill Medical College of Cornell University, New York, NY (Website: