Patient characteristics
A total of 160 patients were initially screened, of which, 2 patients had only 1 type of pathological specimen and 5 patients had only 2 types of pathological specimens; so these 7 patients were eliminated. Ultimately, a total of 153 patients were included in this retrospective study. Among the study group, 122 cases were male, 31 cases were female, and the average age was 63.2 ± 9.2 (41-81). A total of 154 lymph nodes and tumour puncture samplings were analysed, of which, 1 patient had 2 groups of mediastinal lymph node puncture samples. Specific mediastinal lymph nodes and tumour cases are shown in table 1. A total of 119 puncture sites were directly diagnosed as lung cancer by EBUS-TBNA. Only 1 patient in the mediastinal 4R and 7 group of lymph nodes had lung cancer; 118 cases were directly diagnosed as lung cancer, including adenocarcinoma in 34 cases, squamous cell carcinoma in 29 cases, non-small cell carcinoma in 19 cases, poorly differentiated carcinoma in 3 cases, small cell carcinoma in 22 cases, and unable to divide cancer cell types in 11 cases.
Comparison of positive rates of different pathological diagnosis
In the 154 puncture sites, the total positive rate of combination of three pathological treatment types (histopathology, direct traditional smear, liquid-based cytology) was 77.3% (119/154). The diagnosis of positive rates of single pathological treatment were: tissue pathology 68.6% (106/154), direct traditional smear 65.6% (101/154), and liquid base cytology 60.4% (93/154). There was no significant difference among the three single pathological treatment types (p=0.29), but there was a statistically significant difference between the combination of the three treatment types and any single pathological treatment type (p=0.01).
The comparison of positive diagnostic rates of different mediastinal lymph node puncture sites is shown in table 2. Of the 66 punctures near the trachea mediastinal lymph nodes (including 2R, 4R, 4L), the total positive rate of combination of the three types of pathological treatments (histopathologic, traditional direct smear, liquid based cytology) was 87.9% (58/66). The diagnostic positive rates of single pathologic treatment type were as follows: histopathology was 78.8% (52/66), traditional smear was 75.8% (50/66), liquid-based cytology was 71.2% (47/66), and there was no statistically significant difference between the three pathological treatment types (P=0.60). In addition, there was no statistically significant between the combination of three treatment types and any single pathological treatment type (P=0.11).
For 55 punctures for 7 groups (Subcarina) of mediastinal lymph nodes, the total positive rate of combination of three pathological treatment types (histopathology, direct traditional smear, liquid based cytology) is 72.7% (40/55). The diagnostic positive rate for single types of pathological treatment were as follows: histopathology 65.5% (36/55), direct traditional smear 60.0% (33/55), and liquid-based cytology 54.5% (30/55); there were no statistically significant differences (p=0.60) between three single pathological treatment types, and there was no statistically significant correlation (p=0.11) between the combination of three treatment types and any single pathological treatment type
Of 19 punctures in pulmonary hilar lymph node (including 10R, 10L, 11R, 11L, 12R), the total positive rate of combination of three pathological treatment types (histopathology, direct traditional smear, liquid based cytology) was 42.1% (8/19), and the diagnostic positive rates for single types of pathological treatment were as follows: histopathology 31.6% (6/19), direct traditional smear 31.6% (6/19), liquid-based cytology 26.3% (5/19); there were no statistically significant differences between three single pathological treatment types (P=0.92) and there was no statistically significant correlation between the combination of three treatment types and any single pathological treatment type (P=0.77)
Of 14 punctures in paratracheal mass, the total positive rate of combination of three pathological treatment types (histopathology, direct traditional smear, liquid based cytology) was 92.9% (13/14). The diagnostic positive rates for the single type of pathological treatment were as follows: histopathology 85.7% (12/14), direct traditional smear 85.7% (12/14), liquid-based cytology 78.6% (11/14); there were no statistically significant differences between three single pathological treatment types (P=0.84), and there was no statistically significant correlation between the combination of three treatment types and any single pathological treatment type (P=0.75).
Diagnostic efficiency of different specimen treatments
The sensitivity, specificity, positive predictive value, negative predictive value and diagnostic accuracy of a single pathological type for lung cancer diagnosis were as follows: histopathology was 90.6%, 100%, 100%, 77.1%, and 92.9%, respectively, direct traditional smear was 84.9%, 94.3%, 98.1%, 64.7%, and 87.0%, respectively, and liquid-based cytology was 80.2%, 94.7%, 97.9%, 61.0%, and 83.8%, respectively. Comparison of diagnostic accuracy among three single pathological treatment types, there was statistically difference (p=0.04), the diagnostic accuracy of histopathology is the best.
The sensitivity, specificity, positive predictive value, negative predictive value and diagnostic accuracy of the combination of two pathological types for lung cancer diagnosis were as follows: when histopathology combined with direct traditional smear, they were 94.4%, 100%, 100%, 81.1% and 95.5%, respectively; when histopathology combined with liquid-based cytology, they were 92.8%, 100%, 100%, 76.3%, and 94.2%, respectively.
The sensitivity, specificity, positive predictive value, negative predictive value and diagnostic accuracy of the three pathological treatment types combined for lung cancer diagnosis were 96.7%, 100%, 100%, 88.6% and 97.4%, respectively. Comparison of diagnostic accuracy among combination of the three treatment types and combination of the any two treatment types, there was no statistically difference (p=0.62). as shown in table 3.
Diagnostic consistency of the three pathological treatments
Of the 154 puncture sites, the same diagnosis (all positive or negative) were obtained by three methods in 114 cases, accounting for 74.0%. The positive results of traditional direct smear and liquid based cytology were the same, but they do not agree with the result of histopathology in 12 cases, which accounted for 7.7%. Only histopathology and the traditional direct smear had the same positive results in 7 cases, accounting for 4.5%. Only histopathology and liquid based cytology has the same positive diagnosis in 2 cases, which accounted for 1.3%. Only histopathology had single positive results in 16 cases, accounting for 10.4%. There were only 2 cases with single positive results of direct traditional smear, accounting for 1.3%. Only 1 case with single positive diagnosis of liquid-based cytology accounted for 0.6%, as shown in figure 2.