The first clinical application of planning software for laparoscopic microwave thermosphere ablation of malignant liver tumours

Abstract

Background

Liver tumour ablation is an operator-dependent procedure. The determination of the optimum needle trajectory and correct ablation parameters could be challenging. The aim of this study was to report the utility of a new, procedure planning software for microwave ablation (MWA) of liver tumours.

Methods

This was a feasibility study in a pilot group of five patients with nine metastatic liver tumours who underwent laparoscopic MWA. Pre-operatively, parameters predicting the desired ablation zones were calculated for each tumour. Intra-operatively, this planning strategy was followed for both antenna placement and energy application. Post-operative 2-week computed tomography (CT) scans were performed to evaluate complete tumour destruction.

Results

The patients had an average of two tumours (range 1–4), measuring 1.9 ± 0.4 cm (range 0.9–4.4 cm). The ablation time was 7.1 ± 1.3 min (range 2.5–10 min) at 100W. There were no complications or mortality. The patients were discharged home on post-operative day (POD) 1. At 2-week CT scans, there were no residual tumours, with a complete ablation demonstrated in all lesions.

Conclusions

This study describes and validates pre-treatment planning software for MWA of liver tumours. This software was found useful to determine precisely the ablation parameters and needle placement to create a predicted zone of ablation.

Introduction

Although liver resection offers patients with malignant liver tumours the best chance of a cure, the majority of the patients are not candidates because of inadequate liver remnant, underlying cirrhosis and accompanying medical co-morbidities. Thermal ablation modalities, including radiofrequency ablation (RFA) and microwave ablation (MWA) offer treatment options to a significant number of these patients in whom a hepatectomy is not feasible. These ablation modalities are needle-based therapies that rely on the skill and experience of the surgeon or radiologist performing the procedure. The success of the procedure depends on the selection of a correct ablation algorithm, and accurate needle placement, to ensure the creation of an ablation zone encompassing the tumour with a margin.

Pre-treatment planning, with the pre-operative determination of the surgical strategy, in regards to how the lesions in a given patient will be treated, may increase the efficiency and efficacy of the operation. Furthermore, this may also help in simplifying the procedure when an ablation modality with a complex algorithm is utilized.

Microwave thermosphere ablation is a newer technology, which offers significant advantages when compared with RFA, in terms of more efficient tissue heating, with potentially shorter ablation times and fewer treatment failures owing to a less prominent ‘heat sink effect’.^1–8 Although there are reports on pre-planning for percutaneous RFA under computer tomography (CT) guidance in the radiology literature,^9,10 pre-planning for a surgical liver tumour ablation is a new concept, with no reports on MWA. The aim of this study was to describe the initial clinical use of a new, planning software for microwave thermosphere ablation of malignant liver tumours.

Patients and methods

This was an Institutional Review Board-approved prospective study. Pre-treatment planning software (Emprint Procedure Planning Application, Covidien, Boulder, CO, USA) was used to determine the treatment strategy in patients undergoing laparoscopic MWA of malignant liver tumours. The patients’ tri-phasic CT scans were loaded into the software, and the lesions were individually selected using target markers around the tumours in the axial, coronal and sagittal planes. A three-dimensional image of the lesion was created by the software (Fig.1). Using a second set of target markers, the desired zone of ablation was overlapped around the tumour (Fig.2). This projected ablation zone could be moved around the tumour, and its size in three dimensions could be adjusted using the computer mouse. Based on the predicted zone of ablation, the software calculated the ablation diameter, antenna insertion depth, and minimum and maximum margin covered around the tumour. By choosing the power setting, the software calculated the time required to create that ablation zone. Using the computer mouse, the needle angle, insertion length and its position within the tumour could be adjusted for the surgeon to choose the final ablation parameters.

Figure 1.

The computer captures are showing pre-treatment planning, which involves the selection of the liver tumour with the region of interest marker (green circle) in the axial, coronal and sagittal planes. Once this is done, the software creates a three-dimensional image of the lesion. Then another region of interest (red circle) is placed around the tumour, to represent the ablation zone planned. The software also demonstrates how the needle needs to be positioned within the tumour to create this ablation zone. This projected ablation zone can be moved around the tumour, and its size in three dimensions can be adjusted using the computer mouse. The insertion angle and depth of the needle can also be changed. In the final window, the software displays the predicted ablation zone overlapping around the tumour, which can be magnified to see the exact position of the antenna

Figure 2.

The use of the software in a 76-year-old patient with a solitary colorectal metastasis in segment 7 measuring 1.5 × 1.7 × 1.7 cm (a). Initially, target regions of interest are placed around the tumour (green circle). Subsequently, additional targets of interest are placed around the tumour and the planned zone of ablation is created (red circle). This geometry is then applied intra-operatively by laparoscopically placing the antenna inside the lesion as precisely as planned (b) and performing microwave ablation (MWA) at 100 W for 3 min. A post-treatment 2-week computed tomography (CT) scan demonstrates an ablation zone encompassing the tumour, in the shape, size and position predicted by the software (c–d)

This pre-planning software was used and validated in five patients who underwent laparoscopic MWA of nine malignant liver tumours between December 2014 and January 2015, by one surgeon (E.B.). The tumours were treated by following the parameters determined by this software pre-operatively. The microwave probe was inserted under laparoscopic ultrasound guidance into the tumours at the position determined by the pre-treatment software, and the ablation was carried out for the duration of time again calculated by this software pre-operatively. CT scans, obtained 2 weeks after ablation, were analysed for completeness of tumour destruction, guided by the planning software. Continuous data are expressed as the mean ± standard error of the mean.

Surgical procedure

The procedure was done under general endotracheal anaesthesia, with the patients supine. Two grammes of cefazolin was administered intravenously for antibiotic prophylaxis. Two 12-mm trocars were used in the right upper quadrant, 1 for the angled laparoscope and the other for the laparoscopic 10 MHz linear, side-viewing ultrasound transducer (Aloka, Wallingford, CT, USA). Initially diagnostic laparoscopy was performed, followed by surgeon-performed liver ultrasound. In those patients without a pre-operative tissue confirmation, a percutaneous biopsy of a representative lesion was performed using an automated biopsy gun. The size and the location of tumours were recorded. Then the microwave ablation antenna (Emprint Procedure Planning Application, Covidien) was taken into the field and introduced to the abdomen through a 3-mm trocar placed between the ultrasound trocar and the costal margin. The antenna was then inserted into the liver parenchyma parallel to the transducer and positioned within the tumour based on the details obtained from the pretreatment planning software. The ablation was then performed based on the time again that was calculated pre-operatively from the planning software. The needle was then withdrawn out of the liver using ‘track ablation’ where 100 W of power was applied as it was being pulled out. After all the lesions had been ablated, haemostasis was obtained, and the trocars were removed. The fascial incisions for the 12-mm trocar sites were closed with #0 absorbable stitch. This was followed by skin closure. The patients were discharged home on the next day.

Results

There were four females and one male, with a mean age of 66 years (range 56–76). Three patients had colorectal cancer, one patient ovarian cancer and another patient leiomyoma metastases to the liver. The patients had an average of two tumours (range 1–4), measuring 1.9 ± 0.4 cm (range 0.9–4.4). The ablation time was 6.4 ± 1.2 min (range 2.5–10) at 100 W. There were no complications or mortality. The patients were discharged home on post-operative day (POD) 1. At 2-week CT scans, there were no residual tumours, with a complete ablation demonstrated in all lesions. Table1 summarizes the details of each lesion treated in the study.

Table 1.

The details of individual lesions, ablation parameters and ablation zones obtained at 2-week computed tomography (CT) scans

Lesion	Tumour type	Size (cm)	Liver segment	Proximity to <4 mm blood vessel	Location	Ablation power (W)	Ablation time (min)	Predicted ablation Diameter (cm)	Ablation Zone at 2-week CT (cm)
1	Colorectal	3.2 × 2.8 × 1.7	III	Near	Superficial	100	10	4	4.8 × 3.5 × 4.4
2	Ovarian	4.4 × 3.9 × 3.5	VIII	Near	Superficial	100	10	4	3.9 × 4.5 × 3.7
3	Colorectal	0.7 × 0.9 × 0.8	V	Near	Central	100	2.5	3	2.9 × 2.3 × 2.7
4	Colorectal	1.5 × 0.8 × 1.2	V	Near	Central	100	10	4	3.8 × 3.8 × 4.1
5	Colorectal	1.6 × 1.4 × 1.3	VII	Near	Superficial	100	3	3.1	3.2 × 3.4 × 3.4
6	Leiomyosarcoma	1.9 × 1.7 × 1.2	IVb	Near	Superficial	100	4.5	3.4	3.8 × 3.8 × 3.7
7	Leiomyosarcoma	1.1 × 1.2 × 1.1	VII	Away	Superficial	100	10	4	3.8 × 4.3 × 4.3
8	Leiomyosarcoma	1.1 × 1.1 × 0.9	II	Away	Superficial	100	4.5	3.4	3.2 × 3.4 × 3.7
9	Leiomyosarcoma	0.8 × 1.2 × 1.2	IVa	Near	Central	100	3	3.1	4.0 × 3.7 × 4.2

Discussion

This study demonstrates the utility of pre-treatment planning software to determine the ablation parameters for MWA in a pilot group of patients with malignant liver tumours. To our knowledge, this is the first description of planning software for laparoscopic liver tumour MWA. Overall, the software was found to be helpful in outlining the details of ablation for individual tumours before going into the operating room, thereby facilitating the procedure and not leaving the room for unexpected intra-operative decision-making. Subjectively, this increased the efficiency of the procedure. The software was validated by demonstrating complete ablation in all tumours treated, without any complications.

Nevertheless, there are some drawbacks of the software that need to be improved. First of all, it does not allow the surgeon to pick antenna trajectories based on the laparoscopic exposure expected in a given patient. Second, it does not have augmented reality built-in to move the liver in three-dimensions, to again 100% simulate a surgical needle placement. Still, its best utility was recognized as being able to determine ablation parameters for a given lesion, using the new microwave thermosphere technology used in the study. The ablation parameters were more complex compared with RFA and, therefore, the use of this software was found to simplify the ablation planning.

Although, the development of this software is a first step in optimizing ablation parameters, it does not include a needle-tracking feature or automation regarding needle placement. Currently, the performance of liver tumour ablation is operator dependent, with the efficacy of needle placement varying with experience of the surgeon. As shown in other abdominal surgical procedures, robotic assistance can improve the precision of antenna placement. There is ongoing work in automation of this process for percutaneous ablation with robots used in CT scanner units⁹; however, developing an automated laparoscopic needle/antenna placement system for liver tumour ablation involves more complex tasks, such as being able to provide augmented reality and adjusting to different needle angles in a given patient, based on the relationship between the costal margin, liver and visceral organs. Nevertheless, with the developments in robotic surgical systems, future advances are possible in this field.

Solomon et al.⁹ have reported on a percutaneous tumour ablation planning coupled to robotic needle placement. In the system described, predicted ablation zones are overlaid on the CT images of a patient's tumour. However, these images were obtained while the patient was on the table for the procedure. The needle is then loaded into a robotic needle holder. When the accepted coordinates are transferred from the treatment planning system to the robot, the robot manoeuvers the applicator to the correct needle trajectory. The authors validated this technique in two patients, with no evidence of local recurrence at 6 months.

In conclusion, this report describes a novel, pre-treatment software for MWA planning, whose utility was validated in a pilot series of surgical patients. Although it has some limitations, it was found to facilitate the surgical procedures. A prospective study is being conducted prospectively to further analyse the size of the ablation zones achieved in comparison to those predicted by the software, in a larger number of patients.

Conflicts of interest

None declared.