Objective To evaluate the CT features of small lung invasive adenocarcinoma with air?containing space and its relationship with pathological types, and to explore the pathological basis of air?containing space. Methods CT and pathological data of fifty patients with surgically proven lung invasive adenocarcinoma with air?containing space in our hospital from January 2012 to December 2017 were retrospectively reviewed. CT image analysis included image features of tumor and air?containing space. Pathological analysis included pathological subtype, differentiation degree. CT features of tumor and air?containing space were compared with regard to pathological types using chi?square test or Fisher exact text. Analysis of variance was used to compare quantitative data satisfying normal distribution, while those data not satisfying normal distribution were compared with Kruskal?Wallis test. In addition, Spearman correlation was used to analyze the correlation between nodule density and pathological types. Results (1) CT features of tumors: Tumors were predominantly located in peripheral lungs (46/50). Most of the tumors were subsolid nodules (37/50). Tumor?lung interface was generally clear (46/50). Tumors are often accompanied by malignant signs such as lobulation (37/50), spicule sign (27/50), air bronchogram sign (43/50), and pleural indentation (36/50). The mean diameter of nodules ranged from 7.50 mm to 18.12 mm, with an average of (12.91±2.85)mm. The nodule density ranged from-657.00 to 73.00 HU with an average of (-213.88±206.16) HU. (2) CT features of air?containing space:Air?containing spaces were commonly solitary (37/50) and were found to be mainly in an eccentric distribution (29/37). The maximum diameter of air?containing space ranged from 1.00 to 16.00mm, and the average diameter was (4.23±3.14)mm. Air?containing spaces less than 5mm were found in 35 cases (70%), and air?containing spaces more than 5mm were found in 15 cases. (3) Comparison of CT features and corresponding pathological types: Nodule density, number of air?containing space, and type of nodule density in different pathological types were statistically different (P<0.05). There was a correlation between nodule density and pathological subtypes (r=0.371, P=0.008). Differences of nodule density, short?dimension of nodule, type of nodule density, spicule sign, pleural indentation among different tumor differentiation degrees were significant (P<0.05). The degree of tumor differentiation was negatively correlated with nodule density (r=-0.451, P=0.001). Conclusion The detection rate of air?containing space in lung invasive adenocarcinomas is 12.7%. Most small lung invasive adenocarcinomas with air?containing space are presented as peripheral subsolid nodule, and there is a certain correlation between their CT features and pathological types. The pathological basis of air?containing space was supposed to be dilated distal bronchiole induced by check?valve mechanism and destruction of alveolar structure by tumor.

Objective: To evaluate the CT features of small lung invasive adenocarcinoma with air-containing space and its relationship with pathological types, and to explore the pathological basis of air-containing space. Methods: CT and pathological data of fifty patients with surgically proven lung invasive adenocarcinoma with air-containing space in our hospital from January 2012 to December 2017 were retrospectively reviewed. CT image analysis included image features of tumor and air-containing space. Pathological analysis included pathological subtype, differentiation degree. CT features of tumor and air-containing space were compared with regard to pathological types using chi-square test or Fisher exact text. Analysis of variance was used to compare quantitative data satisfying normal distribution, while those data not satisfying normal distribution were compared with Kruskal-Wallis test. In addition, Spearman correlation was used to analyze the correlation between nodule density and pathological types. Results: (1) CT features of tumors: Tumors were predominantly located in peripheral lungs (46/50). Most of the tumors were subsolid nodules (37/50). Tumor-lung interface was generally clear (46/50). Tumors are often accompanied by malignant signs such as lobulation (37/50), spicule sign (27/50), air bronchogram sign (43/50), and pleural indentation (36/50). The mean diameter of nodules ranged from 7.50 mm to 18.12 mm, with an average of (12.91±2.85)mm. The nodule density ranged from-657.00 to 73.00 HU with an average of (-213.88±206.16) HU. (2) CT features of air-containing space: Air-containing spaces were commonly solitary (37/50) and were found to be mainly in an eccentric distribution (29/37). The maximum diameter of air-containing space ranged from 1.00 to 16.00mm, and the average diameter was (4.23±3.14)mm. Air-containing spaces less than 5mm were found in 35 cases (70%), and air-containing spaces more than 5mm were found in 15 cases. (3) Comparison of CT features and corresponding pathological types: Nodule density, number of air-containing space, and type of nodule density in different pathological types were statistically different (P<0.05). There was a correlation between nodule density and pathological subtypes (r=0.371, P=0.008). Differences of nodule density, short-dimension of nodule, type of nodule density, spicule sign, pleural indentation among different tumor differentiation degrees were significant (P<0.05). The degree of tumor differentiation was negatively correlated with nodule density (r=-0.451, P=0.001). Conclusion The detection rate of air-containing space in lung invasive adenocarcinomas is 12.7%. Most small lung invasive adenocarcinomas with air-containing space are presented as peripheral subsolid nodule, and there is a certain correlation between their CT features and pathological types. The pathological basis of air-containing space was supposed to be dilated distal bronchiole induced by check-valve mechanism and destruction of alveolar structure by tumor.