Application of intraoperative laparoscopic ultrasound guidance combined with Sonazoid contrast in surgical treatment of colorectal liver metastasis

Abstract

BACKGROUND

Contrast-enhanced ultrasound (CEUS) performed intraoperatively for colorectal cancer liver metastases can detect concealed liver metastases not identified during preoperative examinations or routine intraoperative ultrasound. This capability is critical for determining the appropriate surgical approach.

AIM

To explore the advantages of Sonazoid-CEUS (S-CEUS) over gray-scale ultrasound (GSUS) for tumor localization during laparoscopic ultrasound-guided microwave ablation (MWA) in colorectal liver metastases (CRLM) and to evaluate the real-time guidance role of S-CEUS in ablation therapy.

METHODS

We retrospectively reviewed data from 259 patients with CRLM who underwent MWA at the First People’s Hospital of Changde City between June 2020 and December 2023. Intraoperative target lesion localization was performed using GSUS or S-CEUS. Target lesion visibility was scored using a 5-point scale, and visibility differences between GSUS and S-CEUS were compared. The duration of the Kupffer phase for the target lesions was observed, and the effects of MWA performed during the Kupffer phase were evaluated.

RESULTS

Preoperative magnetic resonance imaging (MRI) identified 273 lesions in 259 patients, whereas intraoperative S-CEUS detected 281 lesions, all confirmed as liver metastases via needle biopsy. The target lesion visibility scores were 2.87 ± 1.09 during GSUS scanning and 4.07 ± 0.74 (arterial phase) and 4.31 ± 0.58 (Kupffer phase) during the S-CEUS. The visibility scores in both the arterial and Kupffer phases of S-CEUS were superior to those in GSUS (both P < 0.05). The Kupffer phase of target lesions lasted > 1 hour. In patients with multiple pre-known lesions, all lesions were localized with a single contrast agent injection. Three occult lesions undetected via preoperative imaging were identified during the Kupffer phase. All lesions underwent MWA in the Kupffer phase. Immediate reassessment with repeat Sonazoid injection 15 minutes post-ablation and enhanced MRI at 1 month postoperatively demonstrated complete ablation of all lesions.

CONCLUSION

During MWA for CRLM, target lesion visibility in both the arterial and Kupffer phases of S-CEUS was superior to that of GSUS, facilitating improved target lesion localization. The Kupffer phase of S-CEUS provided valuable real-time guidance for MWA.

Key Words: Colorectal cancer; Liver metastasis; Ultrasound guidance; Laparoscopy; Sonazoid contrast imaging

Core Tip: Advances in imaging technology have improved preoperative lesion localization and surgical planning by providing valuable visual references in the surgical treatment for liver metastases from colorectal cancer. However, strategies to better localize occult lesions intraoperatively following chemotherapy warrants further exploration. Sonazoid (Shizhuoan in Chinese) is a contrast agent taken up by Kupffer cells, which could facilitate further localization of hepatic lesions. However, reports on its intraoperative application in China remain limited. This study compared the visualization of target lesions during laparoscopic microwave ablation for colorectal liver metastases using Sonazoid-enhanced contrast-enhanced ultrasound vs gray-scale ultrasound.

INTRODUCTION

Colorectal cancer ranks third globally in incidence among malignancies and second in mortality, making it the leading malignancy of the digestive system in terms of both global incidence and mortality[1,2]. With increasing westernization of lifestyles and dietary patterns in China, the incidence of colorectal cancer has increased, becoming the second most common malignancy by incidence and the most prevalent within the digestive system in China[3]. The liver is the most common site of hematogenous metastasis in colorectal cancer. Approximately, 15%-25% of patients present with liver metastases at diagnosis, and an additional 15%-25% develop liver metastases during disease progression[4]. Colorectal liver metastases (CRLM) is the primary cause of mortality in patients with colorectal cancer. The median overall survival for patients with colon and rectal cancer presenting with liver metastasis at initial diagnosis is 10 months and 13 months, respectively. The prognosis for patients with liver metastasis is worse than those with lung metastasis but better than those with metastasis to other sites[5,6].

Surgical resection remains the most effective option for CRLM[7]. The median survival after surgical resection exceeds 40 months, and no alternative treatment method has achieved comparable outcomes[8]. The survival benefit of surgical resection in CRLM is indisputable[9]. Contrast-enhanced computed tomography (CE-CT) using gadolinium-based extracellular contrast agents and magnetic resonance imaging (MRI) remain the gold standards for preoperative diagnosis of CRLM[10]. Gadoxetate disodium-enhanced (EOB)-MRI enables dynamic evaluation of tumor vascularity in liver tumor differentiation and provides liver-specific imaging through hepatocyte contrast uptake, enhancing preoperative lesion localization and surgical planning with imaging reference value[11]. Nevertheless, effective intraoperative validation methods remain limited. Sole reliance on preoperative imaging may lead to incomplete resection and early postoperative tumor recurrence in some patients. Improved intraoperative localization of occult lesions, particularly after chemotherapy, remains an important clinical challenge.

Intraoperative ultrasound helps address limitations in detecting occult lesions during surgery. Among available techniques, real-time contrast-enhanced ultrasound (CEUS) is particularly effective in identifying liver lesions missed on preoperative imaging in a significant proportion of patients. CEUS can detect concealed liver metastases not identified by preoperative examinations computed tomography (CT) and MRI, or by routine intraoperative ultrasound, which is critical for surgical decision-making[12,13]. CEUS-guided resection has also been associated with reduced positive margin rates in malignant liver tumors[14]. Lesions missed by preoperative MRI or CT are frequently located in anatomical blind spots of the liver, including: (1) The lower segment of the right anterior lobe, adjacent to the diaphragm, where respiratory motion introduces interference MRI artifacts; (2) The upper segment of the right anterior lobe within the diaphragmatic dome, obscured by ribs, leading to reduced CT/MRI resolution); and (3) The liver surface, where small lesions may exhibit signal characteristics similar to normal liver parenchyma on MRI, making them easily overlooked. Sonazoid-CEUS (S-CEUS), through real-time dynamic enhancement and the high-contrast properties of microbubble contrast agents, enhances visualization of abnormal blood perfusion in these regions. Visualization of the diaphragmatic done is significantly improved, particularly when patients are positioned in the sitting or right lateral decubitus position. International research on the clinical application of intraoperative ultrasound for colorectal cancer began earlier. Sonazoid, a liver-specific contrast agent, was approved for marketing in Japan in 2007[15]. Taken up by Kupffer cells in the liver, it enables continuous hepatic visualization during the 5-30 minutes Kupffer phase, facilitating precise hepatic lesion localization. Sonazoid was approved for the Chinese market in 2019; however, reports of its intraoperative application remain limited, and further clinical validation is needed. Therefore, this study compares the intraoperative visualization of target lesions during laparoscopic microwave ablation (MWA) for CRLM using S-CEUS vs gray-scale ultrasound (GSUS). It further analyzes the real-time guidance value of S-CEUS in ablation therapy by observing laparoscopic MWA performed during the Kupffer phase.

MATERIALS AND METHODS

Study subjects

A retrospective analysis was conducted on 259 patients with CRLM who underwent MWA at the First People’s Hospital of Changde City between June 2020 and December 2023.

Inclusion criteria: (1) Age 18-80 years; (2) Colorectal cancer confirmed via endoscopic biopsy or surgical specimen pathology; (3) Postoperative diagnosis of liver metastases based on at least two imaging modalities: CE-CT, CT-MRI, CEUS, EOB-MRI, or positron emission tomography-CT; (4) Acceptable number of metastatic lesions (1-3), suitable tumor location (no involvement of special sites, including tumors protruding beyond the liver surface, proximity to the diaphragm, gallbladder, bile ducts or vessels in the first hepatic portal area, stomach, or transverse colon), sufficient liver reserve [future liver volume (FLR) > 30% in patients without underlying liver disease or FLR > 50% in patients with underlying liver disease], and eligibility for surgical treatment; (5) No invasion of adjacent organs and no portal vein thrombus; (6) Tolerance of general anesthesia with no surgical contraindications; and (7) Complete clinical pathology and imaging data.

Exclusion criteria: (1) ≥ 10 CRLM lesions diagnosed preoperatively; (2) Unresectable tumors adjacent to vital organs or major vascular structures, where safe needle insertion or complete ablation could not be achieved; (3) Extrahepatic metastases, diffuse hepatic metastases, or insufficient residual liver volume postoperatively; (4) Concurrent primary tumors at other sites preventing the identification of the primary site of hepatic metastases; (5) Severe underlying diseases, acute cardiopulmonary dysfunction, acute cerebrovascular disease, or active infection posing high anesthesia risk; (6) Previous treatment of the target nodule, including local ablation therapy or transarterial chemoembolization; (7) Refusal of laparoscopy or general anesthesia; (8) Uncorrectable hepatic dysfunction or coagulation disorders; and (9) Incomplete clinical pathology or imaging data.

The Ethics Committee of the First People’s Hospital of Changde City approved this study.

Target lesion localization

Following successful endotracheal intubation under general anesthesia, routine disinfection and draping were performed, and the patient was positioned in the reverse Trendelenburg position. A 1-cm periumbilical incision was made, and a 10-mm disposable trocar (laparoscopic observation port) was inserted. After establishing pneumoperitoneum, laparoscopic abdominal exploration was performed. Under direct visualization, a small incision was made at the intersection of the right midclavicular line and 2 cm above the umbilical level, and a 12-mm trocar was inserted as the primary working port and intraoperative ultrasound port. At least these two trocars were required for the ablation procedure. An ultrasound technician with more than 10 years of experience scanned the liver lesions using a conventional Doppler ultrasound probe placed on the body surface. An assistant sequentially recorded lesion location, size, and echogenicity. Subsequently, Sonazoid was administered via a peripheral vein at 0.01 mL/kg (1 mL/2 seconds), followed by a 5-10 mL normal saline flush to ensure complete injection. Timing and dynamic image acquisition commenced immediately, with observation lasting more than 10 minutes. The arterial phase (10-30 seconds after Sonazoid injection), portal venous phase (30-120 seconds), and Kupffer phase (> 10 minutes) were recorded separately. A rapid whole-liver scan was conducted to detect localized abnormal contrast enhancement within the hepatic parenchyma. For each detected lesion, images were saved and their locations documented. Lesion location, size, and newly detected lesions were compared with preoperative MRI findings. In cases of multiple or newly suspected lesions, Sonazoid was re-administered until complete lesion documentation was achieved. Target lesion visibility was monitored. Visibility was graded using a 5-point scale: 1: Completely invisible; 2: Almost invisible; 3: Barely visible with poor visibility; 4: Fairly visible; and 5: Clearly visible.

CEUS-guided MWA

The MWA device consisted of a power supply, microwave emitter, computer control system, temperature monitoring system, connecting cables, and cooling system. Under laparoscopic visualization and real-time ultrasound guidance, the puncture site was selected to avoid major vessels and bile ducts. The ablation needle was inserted through the ultrasonic probe guide and advanced along the planned pathway into the deep plane of the target lesion during the Kupffer phase, positioning the needle tip at the anterior edge of the lesion. The MWA device and saline cooling system were activated, with power set to 55-70 W. Single-session ablation time was 2-5 minutes based on tumor size, continuing until significant tumor solidification was observed. The ablation margin extended 0.5 cm beyond the tumor edge to ensure a safety margin. For lesions located < 0.5 cm from the diaphragm, major vessels, gallbladder, and colon, if a 0.5 cm safety margin could not be achieved at the adjacent side but all other margins met the safety requirement, the ablation was considered acceptable. After ablation, the needle was slowly withdrawn with thermal coagulation of the needle tract to reduce bleeding and tract migration. Tumors > 3 cm in diameter underwent multiple ablation sessions to prevent incomplete ablation.

Efficacy evaluation

At 15 minutes post-ablation, S-CEUS was repeated to assess the ablation efficacy. Complete ablation was defined as the absence of focal hyperenhancement in the original lesion area on contrast imaging, with the non-enhancing zone covering the tumor and its surrounding 0.5 cm margin. For tumors adjacent to the liver capsule, vessels, or gallbladder, complete ablation was defined as non-enhancement extending to the tumor’s margin. If incomplete ablation was suspected, additional ablation was performed until intraoperative real-time evaluation confirmed complete ablation. Contrast-enhanced MRI was performed at 1 month postoperatively for re-evaluation.

Statistical analysis

Data analysis was performed using SPSS 25.0. Continuous variables were expressed as mean ± SD. Intergroup comparisons were performed using independent t-tests. P < 0.05 was considered statistically significant.

RESULTS

Target lesion localization

This study included 259 patients. Preoperative MRI identified 273 lesions in these patients, whereas intraoperative Sonazoid CEUS detected 281 lesions, including eight newly identified tumors. The mean tumor size was 2.31 ± 0.41 cm. All target lesions were confirmed as CRLM via needle biopsy (selection of representative lesions). Among the 16 target lesions that were invisible on GSUS due to interference from multiple cirrhotic nodules, nine were well visualized during the S-CEUS arterial phase, and all 16 were well visualized during the S-CEUS Kupffer phase (Figure 1A). Among the 14 target lesions that were not visible during the GSUS phase because they were isoechoic, seven were well visualized in the S-CEUS arterial phase, and 12 were well visualized in the S-CEUS Kupffer phase (Figure 1B). Six target lesions not visible in the GSUS phase due to their size being ≤ 1 cm were well visualized in both the arterial and Kupffer phases of S-CEUS (Figure 1C and D). Table 1 presents the visibility of target lesions across phases.

Figure 1 Images of the target lesion in gray-scale ultrasound phase and S-contrast enhanced ultrasound Kupffer phase or arterial phase. A: The target lesion (indicated by arrow) is isoechoic and invisible in the gray-scale ultrasound (GEUS) phase, but demonstrates good visibility in the S-contrast-enhanced ultrasound (S-CEUS) Kupffer phase; B: The target lesion is isoechoic and invisible in GEUS phase, but demonstrates good visibility in the S-CEUS arterial phase; C: The target lesion is < 1 cm in size, isoechoic, and invisible in the GEUS phase, but demonstrates good visibility in the S-CEUS arterial phase; D: The target lesion is < 1 cm in size, isoechoic, and invisible in the GEUS phase, but demonstrates good visibility in the S-CEUS Kupffer phase.

Table 1 Target lesion visibility across phases.

	Invisible (1-2 points)			Visible (3-5 points)
	Interference from multiple nodules caused by liver cirrhosis	Target lesion isoechoic	Target lesion ≤ 1 cm	Poor (3 points)	Good (4-5 points)
GSUS phase	39	28	14	118	82
S-CEUS arterial phase	-	4	-	54	223
S-CEUS Kupffer phase	-	-	-	17	264

GEUS: Gray-scale ultrasound; S-CEUS: S-contrast-enhanced ultrasound.

Comparison of visibility scores

The target lesion visibility scores were 2.87 ± 1.09 in the GSUS phase, 4.07 ± 0.74 in the S-CEUS arterial phase, and 4.31 ± 0.58 in the S-CEUS Kupffer phase. The visibility scores in both the arterial and Kupffer phases of S-CEUS were superior to those in GSUS. Additionally, the visibility score in the S-CEUS Kupffer phase was superior to that in the arterial phase. All differences were statistically significant (P < 0.05; Figure 2).

Figure 2 Target lesion scores across phases. GEUS: Gray-scale ultrasound; S-CEUS: S-contrast-enhanced ultrasound.

Effects of real-time guidance

The Kupffer phase of target lesions persisted for > 1 hour during the procedure. MWA for all target lesions was performed during the Kupffer phase. Immediate S-CEUS assessment at 15 minutes post-ablation and enhanced MRI at 1-month post-procedure demonstrated complete ablation in all lesions.

Adverse reactions

Three patients experienced decreased blood pressure following contrast agent injection, suspected to be allergic reactions. All recovered rapidly after aggressive treatment for anaphylactic shock, with no subsequent complications or sequelae. No other adverse reactions were observed.

DISCUSSION

Ultrasound is commonly used to guide MWA in the treatment of hepatocellular carcinoma (HCC). Some MWA procedures can be performed percutaneously, whereas HCCs in specific anatomical locations are often treated laparoscopically[16]. However, laparoscopic MWA has certain limitations. In cases involving small tumors, hypoechoic nodules, or cirrhotic background liver, heterogeneous echogenicity is frequently observed. Under GSUS, localization of target lesions may be difficult, particularly when differentiation from surrounding regenerative or dysplastic nodules is required[17]. In this context, ultrasound contrast agents enhance lesion-parenchyma contrast and improve target lesion visibility. Conventional contrast agents (e.g., SonoVue) have a relatively short enhancement window, limiting their guidance role during HCC ablation. Sonazoid, in contrast, is taken up by hepatic Kupffer cells and enters the Kupffer phase approximately 10 minutes following injection. During this phase, sustained parenchymal enhancement is observed, and the Kupffer phase has been reported to last at least 1 hour[18]. Because HCC lacks Kupffer cells, lesions typically exhibit low enhancement or enhancement defects. Therefore, intraoperative S-CEUS theoretically facilitates target lesion localization and provides an adequate time window for surgical resection.

In this study, 13.88% (39/281) of the target lesions were difficult to localize on GSUS due to interference from multiple regenerative nodules in cirrhotic liver. Additionally, 9.96% (28/281) of the target lesions were isoechoic in the GSUS phase, leading to localization difficulties. Small target lesions also posed challenges: Kim et al[19] reported that 47.8% of liver tumors ≤ 1 cm were difficult to detect using GSUS. In our study, 4.98% (14/281) of the target lesions ≤ 1 cm were difficult to localize on GSUS. Difficulty in target lesion localization may increase the risk of incomplete or ectopic ablation. CEUS significantly improves the detection rate and diagnostic accuracy of focal liver lesions. It assists in lesion characterization, grading differentiation of primary liver cancer, and predicting treatment response, thereby supporting therapeutic decision-making[18,20]. However, with conventional CEUS using SonoVue, the enhancement window is short, and repeated injections are often needed for multiple lesions, delaying surgical procedures and increasing technical demands[21]. Sonazoid is a novel contrast agent. Compared to SonoVue, it provides higher-quality ultrasound imaging and may enhance diagnostic confidence. Kupffer phase enhancement also enables preliminary differentiation of liver tumors, providing valuable information clinical diagnosis and management. Owing to prolonged hepatic retention, multiple intrahepatic space-occupying lesions and metastases lesions may be detected in a single session[22]. In this study, eight additional tumor lesions undetected by preoperative imaging were identified using S-CEUS. Compared with intermittent tomographic imaging using CT or MRI, S-CEUS offers real-time dynamic scanning. The Kupffer phase can be repeatedly scanned for more than 1 hour, providing extended windows for repeated imaging. For details, contrast agent injection and arterial phase observation: 5-25 seconds, Kupffer phase assessment: 1-3 minutes, lesion localization and needle placement planning: 10-20 minutes, single lesion ablation: 15-30 minutes, multiple lesion ablation: 20-40 minutes, and immediate post-ablation contrast confirmation: 5-10 minutes. Total procedure duration: Single lesion: 20-40 minutes, 2-3 lesions: 50-70 minutes, and ≥ 4 lesions: 70-90 minutes.

Furthermore, the characteristic washout effect of HCC in the Kupffer phase provides evidence for differentiating benign from malignant occult lesions. For newly identified lesions demonstrating low or no enhancement during the Kupffer phase, contrast-enhanced reperfusion imaging can assist in determining whether the newly detected nodules are neoplastic. Immediate ablation of newly identified lesions can significantly alter the strategy for achieving R0 resection. When additional lesions are located subcapsular, non-perivascular, and < 3 cm in diameter, and malignancy is confirmed via S-CEUS-guided biopsy, intraoperative microwave or radiofrequency ablation may serve as a non-resection R0 strategy, thereby avoiding liver function loss associated with extended resection. Existing studies support immediate intraoperative ablation of micro-lesions detected via S-CEUS, particularly when lesions are multiple, scattered, or located in functionally critical zones, achieving both precise tumor clearance and liver function preservation.

In this study, ablation efficacy was evaluated intraoperatively. Patients with incomplete ablation received re-ablation until intraoperative real-time evaluation confirmed complete ablation. Enhanced MRI at 1 month postoperatively demonstrated no activity in the original tumor areas. Immediate intraoperative real-time evaluation of ablation efficacy is therefore a crucial method for confirming complete tumor ablation. Sonazoid re-injection following MWA enables assessment of tumor lesion status (whether it is completely ablated). Additionally, Sonazoid was well tolerated, with a low incidence of mild adverse reactions. Chou et al[23] reported no severe adverse events among 54 patients receiving Sonazoid; except from one case of abdominal pain lasting 15 minutes, all other adverse reactions were transient and required no medical intervention. In the present study, contrast agents were administered under general anesthesia with endotracheal intubation; therefore, subjective symptoms could not be recorded. Electrocardiographic monitoring revealed no adverse reactions such as arrhythmia. Three patients developed shock immediately after Sonazoid injection, strongly suggesting possible allergic reactions. Because Sonazoid contains a hydrogenated phosphatidylserine shell, caution is recommended when administering it to patients with known allergies to eggs or egg products[24]. None of the three patients had a documented history of such allergies before surgery.

This study has several limitations. First, the study population underwent strict selection and may not represent the broader CRLM population, particularly patients with multiple lesions, unresectable disease, or extensive extrahepatic metastases. The conclusions therefore primarily apply to early-stage metastatic patients who are surgical candidates. Extrapolation to more advanced or complex disease requires caution and consideration of additional evidence. Postoperative multicenter studies involving more diverse populations are needed for validation. Second, detailed descriptions of newly identified lesions was limited. In clinical practice, performing needle biopsies on all S-CEUS-detected lesions is impractical, and the lack of rapid intraoperative pathology support often results in treatment decisions being based on macroscopically visible or imaging-confirmed lesions, while additional nodules may be considered non-critical or technical artifacts and excluded from treatment planning. As of 2026, major international hepatobiliary tumor guidelines—including those from the European Association for the Study of the Liver, National Comprehensive Cancer Network, and European Society for Medical Oncology—do not provide specific management recommendations for additional lesions detected via S-CEUS. Therefore, prospective cohort studies are needed to establish standardized intraoperative documentation and management frameworks for such lesions. Furthermore, no prospective head-to-head clinical studies have directly compared S-CEUS (e.g., Sonazoid) and SonoVue within the same patient population under identical surgical conditions for key parameters. Current clinical inferences suggesting that S-CEUS may be more advantageous are largely based on theoretical differences in the physicochemical properties of the contrast agents rather than on direct comparative clinical evidence. This evidence gap represents a key limitation of the current literature and should inform the design of future high-quality clinical trials. Accordingly, large-sample, prospective, randomized controlled trials are warranted.

CONCLUSION

In summary, during MWA for CRLM, target lesion visibility in both the arterial and Kupffer phases of S-CEUS was superior to that of GSUS, facilitating improved target lesion localization. The sustained Kupffer phase of the S-CEUS, lasting up to 1 hour, enables real-time guidance during MWA.