1. Clinical characteristics of participants:
A total of 472 patients were prospectively included from Nov 1, 2018 to Oct 31, 2019, 12 cases were excluded due to incomplete clinical data and default of follow-up. All clinical data were shown in Table 1. The flow diagram was shown in Fig3
Prior conventional ultrasound showed intra-lesion in 32.17% (148/460) cases and peripheral obvious vessels in 21.30 % ( 98/460) cases , while CEUS presented intra-lesion necrosis in 72.17% ( 332/460)cases and obvious peripheral vessels in 55.43% (255/460)cases of the 460 cases enrolled, Both of which showed significant differences. (P 0.00) Intra-lesion necrosis was more likely to occur in benign lesions ( P 0.00)and intra-lesion obvious vessels were more likely to occur in malignant lesions ( P 0.00) (fig1-2,Table 1).
2. Overall diagnostic efficiency of CEUS-guided pleural biopsy sampling for pleural lesions:
Among 460 cases enrolled,444 cases obtained diagnosis through CEUS-guided biopsy directly, including 313 cases of infectious disease (280 cases of pleural tuberculosis, 10 case of NTM , 23 cases of empyema) and 131 cases of non-infectious disease (129 cases of primary/metastatic pleural malignancy, 2 cases of mesothelioma ) , 16 cases with undefined diagnosis were underwent thoracoscopy, then 100% (16/16) of them got the definite diagnosis of all non-infectious disease (5 cases of mesothelioma , 11 cases of nonspecific pleurisy). Therefore, thoracoscopy identified 5 additional malignant lesions on the base of CEUS-guided biopsy did.
It was worth mentioned that among the 16 cases with undefined diagnosis through CEUS guided biopsy, histopathology indicated malignant lesions in 3 cases, but the specific pathological type could not be determined, which were finally diagnosed as pleural mesothelioma through thoracoscopic surgery sampling.
In 11 cases with a final diagnosis of nonspecific pleurisy, CEUS-guided biopsies obtained the same results as thoracoscopy sampling, and all of these patients were followed up for six months with stable clinical and radiological results.
Thus, we summarized the overall diagnostic accuracy of CEUS-guided pleural biopsy was 98.91 % (455/460): 100% (313/313) of infectious pleural lesions and 96.60 % (142/147) of non-infectious pleural lesions.
Furthermore, in 330 cases of pleural lesions combined pleural effusion, CEUS-guided biopsy increased the diagnostic yield from 60.30 % (199 /330 ) to 98.36 % (325 /330,P 0.00) in all cases, from 15.56 %(14/90) to % 94.44% (85/90) in malignant pleural lesions ( P 0.00)and from 77.08 %(185/240)to 100%(240/240, P 0.00) in infectious pleural lesions.
3. Diagnostic efficiency of CEUS-guided pleural biopsy sampling in infectious lesions:
313 cases of infectious pleural lesions (280 cases of pleural tuberculosis, 10 case of NTM, 23 cases of empyema) were diagnosed through CEUS-guided biopsy. The diagnostic yield of CEUS-guided biopsy in infectious pleural lesions was 100% (313/313), meanwhile the etiology detection rate was 88.82% (278/313).
In 240 cases with pleural lesions and pleural effusion (210 cases of pleural tuberculosis, 7 case of NTM, 23 cases of empyema), CEUS-guided biopsy obtained etiological diagnosis in 205 cases (85.42%, 205/240), which showed significant higher than that of pleural effusion (12.08 %, 29/240, P 0.00). (Table 2)
In 280 cases of pleural tuberculosis, 100% (280/280) were diagnosed through CEUS-guided biopsy using CRS as the reference standard. The etiology detection rate of CEUS-guided biopsy in pleural tuberculosis was 87.50% (245/280). In 210 cases pleural tuberculosis combined pleural effusion, CEUS-guided biopsy increased the diagnostic yield from 95.24 % (200/210) to 100 % (210/210, P 0.78) when using the CRS as the reference standard, meanwhile increased the etiology detection rate from 11.90% (25/210) to 85.24% (179/210, P 0.00).
4. Diagnostic efficiency of CEUS-guided pleural biopsy sampling in non-infectious lesions:
In 147 cases with non-infectious pleural lesions, 142 cases (129 of primary/metastatic pleural malignancy, 2 of mesothelioma, 11 cases of nonspecific pleurisy) were diagnosed through CEUS-guided biopsy. The diagnostic yield of CEUS-guided biopsy in non-infectious pleural lesions were 96.60 % (142/147).
In 136 cases of malignant lesions, 131 cases (129 of primary/metastatic pleural malignancy, 2 of mesothelioma) were diagnosed through CEUS-guided biopsy. The diagnostic yield of CEUS-guided biopsy in malignant lesions were 96.32%(131/136).
In 90 cases non-infectious pleural lesions with pleural effusion, CEUS-guided biopsy obtained histopathological diagnosis in 94.44%(85/90) cases, which showed significant higher than that of pleural effusion ( 15.56%, 14/90 , P 0.00) (Table 3).
In 90 cases of malignant pleural lesions with pleural effusion, CEUS-guided biopsy increased the diagnostic yield from 15.56 %(14/90) to % 94.44% (85/90).
5. Occurrence of adverse reactions
All patients underwent CEUS and USCB. During the sampling process, 23 patients experienced minor complications. The incidence of pneumothorax was 1.1% (5/460), and most of them were small pneumothorax (compression <15%). Only 1 patient with a history of emphysema underwent closed drainage due to pneumothorax compression of 40%. The incidence of local bleeding or sputum blood was 3.91% (18/460) and all of them got improved after short-term local pressure, and no serious adverse events occurred.