The institutional review board of our hospital approved the present retrospective study and waived the requirement for informed consent for collecting data from the related patients. Written informed consent for CT-guided percutaneous localization has been obtained from all patients prior to performing the procedure.
Study subjects
From June 2016 to March 2019, 249 consecutive patients (90 males, 159 females) with 279 pulmonary nodules underwent CT-guided microcoil localization prior to VATS at our radiology department were enrolled into this study. Exclusion criteria for CT-guided microcoil localization were the following: 1) nodules adjacent to the hilum or apparent vascular structure; 2) lesions located in the bronchi; 3) patients who refused microcoil localization or VATS resection.
Of these, 28 patients (male/female: 10/18; mean age: 57.7±8.0 years) underwent simultaneous as above multiple pulmonary nodules (26 two-nodule, 2 three-nodule, 58 nodules in total; Group A), and the remaining 221 patients (male/female: 80/141; mean age: 57.3±11.3 years) underwent localization procedures for 221 nodules (Group B). None of them has received prior needle biopsy examination of the current pulmonary lesion before. The median interval between CT interventional procedure and VATS was 27.8 hours (range, 0.9∽95.7 hours). The clinical characteristics of the patients were summarized in Table 1.
CT-guided microcoil localization
Before CT-guided localization procedures, patients were trained to hold their breath for 5-10 seconds at the end of inspiration. All the planning and localizing CT scans were carried out using 16 or 320-detector-row scanner (Aquilion 16 or Aquilion ONE; Canon Medical Systems, Japan). The following are parameters of the planning CT: scanning method= helical acquisition mode; tube currents= 50mAs; tube voltage= 120kVp; rotation time= 0.5 sec; imaging FOV= 400; slice thickness= 5mm, the following are parameters of the localizing CT: scanning method= axial acquisition mode; tube currents= 50mAs; tube voltage= 120kVp; rotation time= 0.5 sec; imaging FOV= 400; slice thickness= 4mm. The limited slices including lesions and microcoil were acquired in order to reduce radiation dose.
A planning CT scan was performed before percutaneous needle insertion. An appropriate puncture point on a patient’s skin was marked to obtain the shortest needle entry route meanwhile avoiding the inclusion of bullae and vascular structures/vessels. Embolization microcoil (Cook Incorporated, Bloomington, IN 47404, USA) was selected as a localization marker with a wire diameter of 0.018 inches and a length of 7cm. After local anaesthesia with 2% lidocaine, an 18G/10cm percutaneous introducer needle (Cook Incorporated, Bloomington, IN 47404, USA) was used to puncture the skin from the marked point and halted before penetrating parietal pleura.. After confirming the direction of the tip of the puncture needle by the second CT scan, the needle was further advanced into the normal lung parenchyma around the lesions (within 5mm) was carried out. Then the third CT scan was performed to confirm the final position of the tip of the needle before connecting the loading cannula of the microcoil to the needle. Our method for deploying the microcoil was a modified procedure from Powell’s method [5]. The microcoil was implanted adjacent to nodule within 5mm instead of penetrating the nodule, which may induce hemorrhage or inflammation in the nodule and subsequently affect histopathologic assessment [20]. The intention of our method was to leave the proximal end of the microcoil on the visceral pleura, which will be a direct clue for nodule’s position during VATS resection and significantly improve the efficiency of surgery [14].
Basing on the learning curve analysis described by Chao et al [21], the over-33-month (from June 2016 to March 2019) operational process of the interventional radiologists could be divided into two stages: the initial 16-month and the later 17-month. In the initial 16-month, the radiologists had limited experience in evaluating microcoil dislodgement or migration (localization failure) on CT scans and the microcoil implantation used to be only carried out once for each nodule, while in the later 17-month, a repeated localization procedure would be performed once the radiologist the radiologists predicted the proximal end of the first microcoil was likely to be dislodged or detached from the visceral pleura on post-procedural CT images. And the procedure-related complications were evaluated based on the Society of Interventional Radiology Standards of Practice Committee Classification [12].
Surgical procedure
VATS was performed under single-lung ventilation with a double-lumen endotracheal tube and general anaesthesia. The patient was placed in the lateral decubitus position, with the involved lung in the superior location. According to preoperative images, a thoracoscopic port was inserted into the pleural cavity through the appropriate intercostal space. The other two ports were placed to insert a grasping instrument and a linear stapling device.
Under the guidance of VATS and preoperative localization CT (Figure 1a-g), the excision of the complete microcoil and nodule was carefully performed and the specimen was immediately sent for frozen section. The frozen sections were used to evaluate whether an extended resection was necessary and/or the lesion was completely resected. Complete lobectomy was performed unless the lesion was found to be noninvasive cancer, or the patient had an inadequate cardiopulmonary reserve, or the patient had lung resection before, or the patient declined to extended resection.
Assessment
The successful targeting rate, localization rate, VATS rate, and procedure-related complications rate were calculated based on the total number of the nodules [11,22]. Successful targeting was defined as implantation of microcoil at the target site adjacent to a nodule on CT image which was obtained immediately after the marking procedure and the rate was calculated as follows: (number of successful targeting procedures/number of all localization procedures in each group)x100; successful localization was defined as detection of nodule location and the rate was calculated as follows: ([number of successful targeting procedures-number of dislodgements or misses under the thoracoscope]/number of all localization procedures in each group)x100; successful VATS was defined as a complete resection of the target nodule with adequate margin and the rate was calculated as follows: ([number of successful VATS /number of all localization procedures in each group)x100. The severity of procedural complications was also recorded according to the Society of Interventional Radiology Standards of Practice Committee classification of complications [12].
In addition, all preoperative CT data was transmitted to the picture archiving and communication system (PACS) and scanner workstation. Nodules were classified as solid, part-solid and ground-glass opacity (GGO) according to their density on CT images with a lung window setting (level:-450HU; width: 1300HU). A nodule is defined as GGO when it has increased attenuation relative to lung parenchyma but has not as dense as soft tissue (such as the parenchymal vessels). Part-solid nodule contains some areas with solid attenuation. The nodule characteristics and localization procedure-related variables, including nodule location, size, type, depth from pleura (the shortest vertical distance), presence of emphysema (around the needle insertion pathway), patient position for localization procedure (prone or supine), needle-pleural angle, pleura-microcoil distance (along the needle insertion pathway), presence of “pleural indentation” (defined as the manifestation of the pleura which is not penetrated by the needle but protruding towards the nodule, resulting in a tent-shaped appearance of the pleura), scapulae-covered sign (where nodule is shadowed by the scapulae), presence of procedure-related complications (pneumothorax or pulmonary hemorrhage), localization procedure time (defined as the interval time between the initial CT scan scout film before puncture and last CT scan following completion of localization procedure) as well as the time to the operation (defined as the interval time between the termination of the postprocedural CT scan in the interventional unit and the start of the general anaesthesia in the operating room) were measured and recorded. The comparison of clinical characteristics and procedure-related variables between Group A and B were recorded in Table 1 and 2.
Statistical analysis
Statistical analysis was performed using SPSS 17.0 software (SPSS 17.0 for Windows, Chicago, IL). The Kolmogorov-Smirnov test for normality was performed on continuous variables and the graphical spread of the data was visually inspected. Descriptive statistics were shown as mean ± standard deviation (SD) or median± interquartile range (IQR) for continuous variables, and as frequency and percentage for categorical variables.
The comparison of clinical characteristics and microcoil localization procedure-related variables between group A and group B were analyzed by Independent-samples t-test/Mann-Whitney U test, and the chi-square test/Fisher exact test. A two-sided p-value less than 0.05 was considered statistically significant.