Preoperative ultrasound-guided dual localization with titanium clips and carbon nanoparticles for predicting the surgical approach and guiding the resection of Siewert type II esophagogastric junction adenocarcinoma

This study included patients who were diagnosed with AEG based on endoscopic biopsy and pathological examination at the Ninth Hospital of the Joint Logistics Support Force. Based on the findings of electronic gastroscopy, gastrointestinal barium meal, abdominal CT examination, electrocardiography, cardiac ultrasonography, lung function tests, and other examinations, a comprehensive evaluation was conducted to determine whether the patients were suitable for curative surgery. Only those who were eligible for surgical procedures were considered for the study. A total of 66 patients were selected and randomly divided into an experimental group and a routine group comprising 33 cases each. The inclusion criteria were as follows: (1) diagnosis of AEG based on the pathological results of endoscopic biopsy, (2) classification of AEG as Siewert type II, (3) diagnosis of cIA–IIA stage disease based on CT and EUS examination, (4) no history of other malignant tumors or tumor recurrence and ability to tolerate surgery based on the comprehensive evaluation mentioned earlier. The exclusion criteria were (1) diagnosis of cIIB–IV stage disease based on CT and EUS examination, (2) tumor metastasis that makes resection impossible, (3) staining contamination or detachment of carbon nanoparticle or titanium clip, (4) a history of spinal deformities or fractures, (5) poor cardiopulmonary function and resultant inability to tolerate surgical trauma, and (6) presence of combined malignant tumors. Figure 1 illustrates the patient flow through the trial.

This study was conducted in accordance with the Declaration of Helsinki and was approved by the Ethics Committee of the 900th Hospital of the Joint Logistics Support Force of the PLA (Approval no. 2021–047).

The preoperative clinical staging of the patients was done according to the 8th edition of the AJCC cancer staging system. After patients in the experimental group had undergone routine examinations such as electronic gastroscopy, pathological examination, gastrointestinal barium meal, and abdominal CT examination, dual localization with titanium clips and carbon nanoparticles was performed under direct EUS 1–2 days before surgery. The patient were fasted for at least 4–6 h before labeling. After induction of anesthesia, endoscopists perform EUS to examine the esophagus, esophagogastric junction, stomach, and surrounding lymph nodes. Through direct observation (Fig. 2A) and ultrasound detection (Fig. 2B), the actual depth of tumor infiltration and the actual height of invasion of the esophagus were determined. Two points located 1 cm from the oral side of the actual boundary of the tumor were selected, and 0.1 ml of a carbon nanoparticle suspension was injected into the protruding esophageal or gastric wall of the two points. Next, two additional points at the same height and on the same plane as the previous injection points were selected and marked with titanium clips. It was ensured that the carbon nanoparticle suspension was free of contamination, the injection point was free of bleeding, and the titanium clips were not dislodged (Fig. 2C and D). After the patient had woken up, abdominal radiography was performed to confirm the height of the titanium clip (Fig. 2E, F, and G). The routine group patients underwent routine examination with electronic gastroscopy, pathological examination, gastrointestinal barium meal, and abdominal CT examination, without the use of titanium clips and carbon nanoparticle positioning markers.

In the experimental group, the surgical approach was determined based on the positional relationship between the titanium clip in the abdominal standing plain radiograph and the 10th thoracic vertebra: a transthoracic approach (TTA) was adopted upstream of the 10th thoracic vertebra; a transabdominal approach (TAA), below the 10th thoracic vertebra; and a combined thoracoabdominal approach (CTA), between the upper and lower edges of the 10th thoracic vertebra. During the procedure, the resection range of the lesion was determined by palpating the titanium clip and observing the range of black staining with the carbon nanoparticles (Fig. 3A, B, C, and D).

In the routine group, after comprehensive evaluation based on abdominal radiography, abdominal CT examination, and other exams, either transabdominal surgery or transabdominal surgery was performed. If the surgery could not be completed due to difficulties in anastomosis or positive results for the intraoperative frozen pathological samples, combined thoracoabdominal surgery was performed. Lymph node dissection includes D2 lymph node resection, middle and lower mediastinal lymph node dissection, and bilateral lymph node resection (Moehler et al. 2015).

After the specimen was detached, it was cut along the greater curvature of the esophagus and stomach to keep a safe margin around the lesion and avoid cutting into the lesion itself. It was then unfolded and placed on a piece of coordinate paper. The length of the proximal esophageal part of the resected specimen was measured using a graduated ruler (Fig. 4A). After measurement, the specimens were fixed with 10% formalin fixative. After 24 h, additional specimens were obtained from the visible boundary, with an interval of 0.3 cm, until the upper edge. Each tissue was preserved in a wax block, and sections of the wax block were used to make pathological slides. The presence of tumor cells was observed under an electron microscope, and the extent of tumor infiltration was calculated based on the observations made from the slide (Fig. 4B).

All data were statistically analyzed using SPSS 20.0. Categorical variables were expressed by counts and percentage, and were compared between the two groups using Chi-square tests or Fisher tests. Continuous variables were expressed as mean ± standard deviation and were compared between the two groups using the t-test. P < 0.05 was considered to indicate statistical significance.

Between September 2021 and September 2023, a total of 66 patients were randomly divided into an experimental group (n = 33) and a routine group (n = 33). The patients in the experimental group underwent radical surgery after labeling with titanium clips and carbon nanoparticles, while the patients in the routine group underwent radical surgery without labeling. After surgery, two patients were excluded from the experimental group: one patient was excluded because of contamination of the carbon nanoparticle staining suspension, and one patient was excluded because of peritoneal metastasis. In the routine group, four patients were excluded: one patient had an unresectable tumor, two patients had peritoneal metastasis, and one patient withdrew from the study. After these patients were excluded, 60 patients were included in the final analysis, including 31 patients in the experimental group (24 males and 7 females; average age, 63.43 ± 7.28 years) and 29 patients in the routine group (24 males and 5 females; average age, 61.25 ± 5.66 years). There were no significant differences in the baseline characteristics of the patients, with the exception of the distribution of tumor sites, surgical methods, and reconstruction methods (Table 1).

After the evaluation was completed, the surgical approach was adjusted according to the actual situation during the operation. The accuracy of prediction of the surgical approach was calculated as the number of correct cases divided by the total number of cases in the group. The results showed that the accuracy of prediction for the experimental group and the routine group was 96.8% (30/31) and 75.9% (22/29), respectively (P = 0.02). The values imply that the accuracy of prediction was significantly higher in the experimental group than in the conventional group (Fig. 5, Supplementary Table S1).

The esophageal resection length of the experimental group and the routine group was 2.39 ± 0.28 cm and 2.86 ± 0.39 cm, respectively (P < 0.001), and the length of tumor infiltration into the esophagus was 1.15 ± 0.22 cm and 1.07 ± 0.16 cm, respectively (P = 0.12). The frozen pathological specimens from both groups were negative for tumor cells (Supplementary Table S2). These results confirm the widespread presence of infiltrating Siewert type II tumors in the entire cohort. Based on the results in the experimental group, it appears that it is possible to resect a shorter esophageal segment while ensuring a negative upper margin with the dual localization technique for prediction of the optimal surgical approach and delineation of the resection margin during the procedure (Fig. 6).

The overall incidence of complications in the experimental group and the routine group was 22.59% and 24.14%, respectively (P = 0.88). In the experimental group, there were three cases of pulmonary infection, one case of abdominal infection, two cases of incision infection, and one case of anastomotic stoma fistula. In the routine group, there were four cases of pulmonary infection, two cases of incision infection, and one case of anastomotic stoma fistula. The incidence of postoperative complications in the experimental group was not significantly higher than that in the routine group and that reported in similar studies(Mine et al. 2022) (Fig. 7, Supplementary Table S3).