Ethics statement and patients
All procedures were carried out in accordance with the principles of the Declaration of Helsinki. The study was approved by the Clinical Research Ethics Committee at the Fujian Provincial Hospital, Fujian Province, China (K2015-022-01, September 30, 2015). Informed consent was obtained from the patients included in the study.
Inclusion and exclusion criteria
Patients whose chest CT scans showed pulmonary GGNs and who underwent segmentectomy at Fujian Provincial Hospital between March 2016 and September 2018 were enrolled. The inclusion criteria were as follows: a single lesion located in the lateral one-third of the lung parenchyma with a diameter of ≤2 cm, and according to the indications for segmental resection in the National Comprehensive Cancer Network (NCCN) guidelines, a uniportal video-assisted thoracoscopic segmentectomy was indicated for the patients. The exclusion criteria were as follows: central lesions with a tumour-to-intersegmental fissure distance of <2 cm; patients who do not meet the indications for segmental resection stipulated in the NCCN guidelines; and general conditions that do not allow surgery or preoperative examinations that suggest distant metastasis.(see study flow diagram)
General information
All patients admitted to the hospital for surgical treatment, excluding those with poor surgical tolerance, distant metastasis, stroke sequelae, severe lung ventilation dysfunction, and other conditions, routinely undergo pulmonary function tests, cardiac ultrasound, total abdominal colour Doppler ultrasound, cranial magnetic resonance imaging, and whole-body bone scanning. All patients who met the enrolment criteria were included and randomly divided into experimental group A and control group B using the coin method. For the experimental group A, 3D chest CT reconstruction combined with the 3D printing technique was used to simulate the model to analyse the anatomical relationship and variations. For control group B, only a 3D CT scan was performed. A total of 89 patients were included, with 51 in the experimental group A [21 men and 30 women, aged 43-80 years, mean age 62 years; 17 patients were smokers (all men)] and 38 cases in the control group B [14 men and 24 women, aged 40-81 years, mean age 61 years; 9 patients were smokers (all men)]. There were no significant differences in the general characteristics between the two groups (Table 1).
3D reconstruction and 3D printing
All patients in both groups underwent plain and enhanced chest CT scans for further diagnosis and to determine tumour location. 3D reconstruction and 3D printing were further performed for the experimental group to further locate the tumour, determine the anatomical relationship and any anatomical variations, and to simulate the operative process. The Siemens Sensation 64-slice CT scanner with 1.2-mm pitch and 1.0-mm scanning thickness. The contrast agent used was iofol, which was manufactured by Jiangsu Hengrui Pharmaceutical Co. Ltd. (100 ml: 74.1 g Chinese medicine quasi-word H20143027) and administered using an intravenous injection to the elbow. The arterial and venous phase images were collected 25 and 55 seconds after the injection of the contrast agent, respectively. The IQQA®-Chest system was used to preserve the pulmonary artery, pulmonary vein, bronchus, tumours, hilum, and swollen lymph nodes and to reconstruct these structures with a 1.50-mm thickness. The pulmonary artery was red, the pulmonary vein was blue, and the bronchus was white. The 3D model was printed with Objet1000 Plus (STRATASYS Company). The physical printing ratio was 1:1, and the printing material used was photosensitive resin (Figure 1).
Preoperative positioning
The “coordinate positioning method” was utilised for preoperative positioning. This method is widely used for mapping and determining object positions. Generally, to determine the position of a point, the number or angle should be identified. In the plane, two axes perpendicular to each other and having a common origin to form a plane rectangular coordinate system are identified, with the horizontal and vertical axes noted as such. This method is somewhat different when determining the location of the lesion in the experimental and control groups.
The horizontal axis data are derived from the measurement of the horizontal level of the lung lesion seen in the CT image. The ‘clock alignment method’ is adopted, assuming that the horizontal plane of the right chest is a clock and the lesion is the part indicated by the hour hand. Using the right upper lung lesion as an example (Figure 2A), the horizontal CT section where the lesion is located is selected, the right chest contour is regarded as the clock face, the midline of the clavicle is the 12 o'clock position and the right midline is intended to be the 9 o'clock position (Figure 2B). According to the ‘clock positioning method’, the lesion is at 7.5 points. In both groups, the horizontal axis clock positioning method can be established in the CT horizontal plane.
The measurement of the vertical axis adopts a ‘scale-localisation algorithm’. There was a difference in the proportional positioning method between the experimental and control groups. In the experimental group, the upper end to the lower end of the lobe where the lesion is located on each longitudinal axis is the measurement interval. Depending on the lobes, the upper end can be the tip of the lung or the interlobular fissure, whereas the lower end can be the interlobular fissure or the base of the lung. On the 3D lung model, the distance from the lesion to the upper and lower ends was measured on the vertical axis, and the proportional position of the lesion on the vertical axis was calculated. Continuing to take the right upper lung lesion as an example, the axis on the 7.5-hour position on the 3D lung model was selected. The length of the lesion to the tip of the lung and to the interlobular fissure is recorded as ‘a’ and ‘b’, respectively. The proportional position of the lesion on the vertical axis was calculated using the formula [a/(a+b)] (Figure 2C).
In experimental group A, the vertical position was located using the proportional segmentation method with the scale positioning technique, which was performed on the 3D lung model. The steps are as follows: (a) locate the lobe where the nodule is located on the 3D model; (b) measure the distance between the nodule and the upper boundary of the lobe (a) and the length of the upper and lower boundaries of the lobe (a+b); (c) the ratio of the distance from the upper boundary (a) to the length (a+b) of the upper and lower boundaries of the lung lobe [a/(a+b)] is the position of the nodule in the longitudinal axis of the lobe.
In control group B, preoperative vertical positional localisation was performed according to the sagittal and coronal planes of the CT image, using the following steps: (a) identify the lobe where the nodule is located on the CT image; (b) count the number of CT slices of the nodule from the border of the lung (n) and the total number of CT layers (N) of the upper and lower boundaries of the lobe; (c) calculate the ratio of the number of CT layers (n) of the nodule to the upper boundary of the lobe to the total number of CT layers (N) of the upper and lower boundaries of the lobe, thereby positioning the nodule longitudinally in the lobe.
In addition to the coordinate localisation method, we also selected anatomical landmarks such as the apex of the lung, bottom of the lung, front edge of the lung rib, midline of the lung rib surface, posterior edge line of the lung rib surface, and the interlobular fissure as references. Maneuverer, the lung segment area, tracheal block expansion, and finger touch detection methods, among others, were performed to position the lesion accurately.
Surgical methods
All patients underwent uniportal video-assisted thoracoscopic segmentectomy and systemic lymphadenectomy. The operation was performed by the treatment team led by the chief physician of the Department of Thoracic Surgery, Fujian Provincial Hospital. The assistant adopted a “same side, high position, single hand, sideway” posture mirror [6] and the operator performed the procedure using a thoracoscopic instrument. The location of the intraoperative nodules was determined using data from the preoperative positioning methods described previously. A wedge-shaped resection was initially performed and followed by an intraoperative rapid frozen pathological examination. According to the rapid freezing pathological examination, the segmental vein, artery, and bronchus were separated at the anatomical level and the linear cutting suture device was broken or disconnected. To determine the inter-segment plane, the ‘Lung Expansion-Falling Method’ [7] was utilised for the last segment of the lung fissure. The two groups were separated from the inter-segment plane to the rib surface, and a straight-section cutting stapler was used to process the inter-segment plane. During the operation, the frozen tumour and margin tissues were sent for pathological examinations and the combined lung segment or lobectomy was determined according to the distance between the tumour and the margin (2 cm). Then, a conventional systemic lymph node dissection was performed. One chest tube was placed in the posterior margin of the incision, and one micro-thoracic tube was placed in the lower thoracic cavity (Figure 3).
Observation indicators
The surgical method conversion rate, operative time, intraoperative blood loss, postoperative complication rate, and segmental conversion rate of the lobectomy were measured and the differences between these variables were compared between the two groups. Surgical transfer rate is difficult to complete owing to pleural adhesion. Similarly, microscopic resection and laparoscopic mass resection are difficult to complete and the number of cases that undergo a successful transthoracic surgery is limited due to intraoperative bleeding. The surgical method conversion rate is the proportion of cases in which the segmental resection was converted to a lobectomy due to various factors during the operation. The operative time was defined as the time from the start of the skin incision to suturing completion (h). Intraoperative blood loss is defined as the amount of blood absorbed by the gauze and the intraoperative suction (ml).And it is determined using the weighing method to calculate the attracting liquid hemoglobin content.
Statistical analysis
All data were analysed using SPSS21.0 statistical software. Measurement data were expressed as and the two-sample t-test was used for comparison between groups. The number of count data was used (n) and the ratio of count data (%) was calculated by the X2 test or Fisher’s exact probability method. P<0.05 indicated a statistically significant difference.