The limitations of laparoscopic surgery include the loss of tactile feedback and the inability to palpate deep parenchymal lesions, which may impair intraoperative surgical assessment. In humans, the increasing use of laparoscopic ultrasound (LUS) has helped to overcome these obstacles, as LUS provides real-time evaluation of intraparenchymal structures that are unable to be visualized and palpated during laparoscopic surgery.1,2 The clinical use of LUS in dogs is limited to a single descriptive case series in which lesions identified by laparoscopy were then evaluated via LUS. In 3 dogs, biopsies were obtained from the liver using LUS guidance, and all biopsies were determined to be diagnostic.3
To date, laparoscopic surgery of the liver in dogs is mostly limited to hepatic biopsies, but ablation procedures and peripheral liver resections have recently been described.4,5 Laparoscopic liver resection and ablation are not widely performed in dogs, despite the many known benefits of laparoscopic surgery.6–8 In people, laparoscopic liver resection and ablation procedures are frequently performed. LUS is often used in conjunction to allow oncologic margin evaluation, visualization of surrounding vasculature, and accurate placement of ablation probes.1,2,9,10 The use of LUS as an adjunct to laparoscopic liver surgery in dogs may also be beneficial. Therefore, evaluation of LUS of the liver and technical descriptions are indicated in dogs.
The primary objective of this study was to describe the feasibility and technique for LUS of the liver in healthy dogs. A secondary objective was to report limitations and complications as well as the technical demand associated with LUS of the liver. We hypothesized that LUS of the liver would be feasible and safe but that imaging of the right division of the liver may be more challenging. Additionally, we predicted that the technical demand and ease of performing LUS would improve over time.
Methods
Ethics
This study was approved by the University of Florida IACUC and the Small Animal Hospital Board at the University of Florida (IACUC No. 02111537). Written informed consent was also obtained from all pet owners before study participation.
Case selection
Twelve, healthy, client-owned dogs were prospectively enrolled from January 1, 2022, to October 31, 2022, at a single academic institution (Small Animal Hospital, College of Veterinary Medicine, University of Florida). Dogs were divided into 2 groups based on their weight with 6 dogs weighing ≥ 20 kg and 6 weighing < 20 kg. Each dog underwent a complete general physical examination, as well as CBC and biochemistry before anesthesia and surgery. Inclusion criteria were dogs presenting to the institution for elective laparoscopic surgery including ovariectomy, gastropexy, or a combination thereof. Dogs were excluded from the study if they were determined to be greater than American Society of Anesthesiologists grade 2 based on physical exam as well as CBC and biochemistry results or if they had known preexisting liver disease based on previous bloodwork or imaging diagnostics.
Surgery
All dogs were anesthetized according to standard clinical protocol according to a board-certified veterinary anesthesiologist. Following induction of general anesthesia, dogs were placed in dorsal recumbency, and their ventral abdomen was clipped and aseptically prepared. All dogs received peri-operative cefazolin 22 mg/kg, IV, up to 30 minutes before the first incision, and every 90 minutes thereafter.
Dogs were positioned in dorsal recumbency at approximately 30° reverse Trendelenberg on a hydraulic surgical float table (8000 HLT X-Ray Table; Durabuilt Medical) as previously described for laparoscopic access to the liver.4,5 All laparoscopic exploration and LUS procedures were performed by the same board-certified surgeon (JBC), who has extensive experience with laparoscopic surgery. A 30-mm single incision laparoscopic surgery port (SILS Flexible Port; Medtronic) was placed on the midline centered over the umbilicus in dogs ≥ 20 kg or caudal to the umbilicus in dogs < 20 kg. The abdominal cavity was insufflated to a pressure of 8 to 10 mm Hg. A 5-mm, 30° laparoscope with a high-definition camera (KARL STORZ Endoscopy-America) was placed in the port. The primary surgeon (JBC) was positioned on the right side of the dog for imaging and exploration of the left division and on the left side of the dog for imaging and exploration of the right and central divisions of the liver. The endoscopy tower was positioned on the right side of the dog, with video monitors located over the cranial end of the dog. The LUS machine was positioned on the left side of the dog, below the laparoscopic video monitors, and a nonsterile assistant (FPS) operated the machine (Figure 1). The individual operating the ultrasound (US) machine was a graduate student with 1 year of experience practicing veterinary medicine at the time of study completion.
Laparoscopic exploration of the liver was then performed using the 30° laparoscope to visualize each liver lobe as previously described, with 45° lateral rotation of the dog in the right and left direction for evaluation of the left and right divisions, respectively.5 A 5-mm blunt laparoscopic probe was used for manipulation of the liver lobes or other interfering organs to aid in visualization. All liver lobes were assessed; the ability to completely visualize each lobe in its entirety, including the hilus, was subjectively determined; and images of each lobe were acquired. Any gross abnormalities appreciated were recorded and described, and images were acquired and saved with the laparoscope. Time to perform the laparoscopic exploration was recorded, starting from the moment that the laparoscope was introduced into the port, and ending with exploration of the last liver lobe.
LUS was then performed using a US monitor (ProSound alpha-7 Surgical Ultrasound System; Hitachi Aloka Medical) and a 13- to 5-mHz curved array laparoscopic 4-way transducer (UST-9150 4-way Laparoscopic; Hitachi Aloka Medical; Figure 2). The tip of this transducer can be articulated approximately 180° in the sagittal and frontal planes, and the diameter is 10 mm. Before surgery, the transducer was sterilized with 0.55% orthophthalaldehyde solution (CIDEX OPA; Advanced Sterilization Products) at 20° C for at least 12 minutes. Immediately before use, the transducer was rinsed with sterile saline. A sterile sheath (Camera Sheath; DeRoyal) was used to cover the cord and handle of the transducer. The transducer was inserted into a 12- or 15-mm cannula placed in the single incision laparoscopic surgery port (Figure 3), and each liver lobe was scanned by placing the transducer directly on the ventral parenchymal surface. The probe was moved in a side-to-side manner from the hilus to the apex of each lobe. Scanning started at the left lateral lobe and ended with the caudate lobe. The ability to access each lobe to the level of the hilus with the laparoscopic transducer was subjectively determined. Any lobes that were not able to be accessed with the laparoscopic transducer were recorded. The same lateral rotation as was previously described for access to the right and left divisions was performed. The primary surgeon stood on the left side of the dog to scan the right and central divisions of the liver and stood on the right side to scan the left division. Ultrasonographic abnormalities were recorded and described, and images or cine loops were acquired and saved. Time to perform LUS of the entire liver was recorded, starting from the moment that the transducer was introduced into the port and ending with the completion of scanning the last liver lobe.
Finally, all dogs underwent their intended elective procedure (ovariectomy, gastropexy, or both) routinely. Additional ports were placed as needed to complete the procedure, and the number and location of these ports were recorded. Following the completion of surgery, port incisions were closed routinely. Postoperatively, dogs were managed with gabapentin at 10 to 20 mg/kg, PO, q 8 h, for 14 days; trazodone at 3 to 5 mg/kg, PO, q 8 h, as needed for anxiolysis for 14 days; and carprofen at 2.2 mg/kg, PO, q 12 h, or meloxicam at 0.1 mg/kg PO, q 24 h, if indicated based on bloodwork for 5 days. Dogs returned in 10 to 14 days for incision assessment and suture removal if skin sutures were placed.
Any intraoperative complications that occurred were recorded and graded according to the Clavien-Dindo scheme.11,12 Immediately following the completion of surgery, the primary surgeon (JBC) completed the National Aeronautics and Space Administration Task Load Index (NASA-TLX) questionnaire for the LUS procedure.13 This is a previously validated questionnaire used for the evaluation of technical demand and learning curves that has been applied to many fields, including both human and veterinary laparoscopic surgery.14–17 An overall score from 0 to 100 (0 being least difficult, 100 being most difficult) is determined by a weighted average of 6 categories including mental demand, physical demand, temporal demand, effort, and perceived performance.
Statistics
Descriptive statistics were performed using commercially available statistic software (SAS version 9.4; SAS Institute). Categorical variables were summarized by frequencies and percentages. The Shapiro-Wilk test was performed for all numerical, continuous variables to assess for normality. For variables following a normal distribution, they were summarized with mean, range, and SD. Continuous variables that did not follow a normal distribution were summarized with median and IQR. Nonparameteric data were compared using the Mann-Whitney U test, which was performed to compare the NASA-TLX score and the time to perform LUS between dogs weighing < 20 kg and those weighing 20 kg and above.
Spearman’s rank correlation was performed to assess if there was a relationship between NASA-TLX score with each trial performed, as well as time to perform LUS with each trial performed. P < .05 was used to determine significance.
Results
Signalment and clinical data
Twelve dogs were prospectively enrolled that met the inclusion criteria. Mean body weight was 25.89 kg (range, 5.7 to 62 kg; SD, ± 17.15 kg), with 6 dogs weighing ≥ 20 kg (mean, 40.18 kg; SD, ± 11.83 kg; range, 29.7 to 62 kg) and 6 weighing < 20 kg (mean, 11.6 kg; SD, ± 4.1 kg; range, 5.7 to 16.3 kg). Breeds represented included German Shepherd dog (n = 2), Border Collie (2), mixed breed dog (1), Bernese Mountain Dog (1), Pembroke Welsh Corgi (1), Neapolitan Mastiff (1), Shetland Sheepdog (1), Australian Shepherd (1), Labrador Retriever (1), and Bichon Frise (1). Median age was 1.96 years (IQR, 1.46 to 3.25 years). Sexes were intact female (n = 8), spayed female (1), castrated male (2), and intact male (1).
Laparoscopic exploration
Mean time to perform laparoscopic exploration of the liver was 4.6 minutes (range, 2 to 7 minutes; SD, ± 1.5 minutes). Gross abnormalities were observed in 1 dog who was the oldest study participant (8 years) and had small (1 to 2 mm) multifocal pale areas on the liver surface of the right medial, quadrate, and left medial lobes.
A complication occurred in 2 dogs during laparoscopic exploration. When using the blunt probe to manipulate and mobilize the right lateral liver lobe a small capsular laceration was created in the caudate process of the caudate lobe of 1 dog and in the right lateral lobe of another dog. In both dogs, this resulted in mild, self-limiting venous hemorrhage and did not require any surgical or medical intervention and was therefore considered Clavien-Dindo grade I. Hemostasis was confirmed before closure, and both dogs recovered uneventfully.
All liver lobes were able to be completely visualized to the level of the hilus in 11/12 (91.67%) dogs: in 1 dog the caudate lobe was unable to be completely visualized due to the duodenum and right lateral liver lobe obstructing its view.
LUS
Median time to perform LUS of the liver was 9 minutes (IQR, 5 to 16.5 minutes). Median time to perform LUS was not significantly different between dogs weighing < 20 kg (median, 8.5 minutes; IQR, 8 to 9 min) and those weighing 20 kg and above (median, 15.5 minutes; IQR, 9 to 17 minutes; P = .17). There was a strong negative correlation between time to perform LUS and trial number (r = −0.77; P = .0037). Ultrasonographic abnormalities were noted in 2 dogs. One had a mild amount of gravity-dependent echogenic debris within the gallbladder, and another who had numerous cholecystoliths. No ultrasonographic abnormalities were appreciated in the dog with gross pale areas within the liver parenchyma. No complications occurred during LUS.
All liver lobes were able to be scanned entirely to the level of the hilus in 10/12 (83.33%) dogs (Figures 4, 5, and 6). In 2 dogs, the caudate lobe was unable to be completely scanned due to obscurement from the duodenum. In 1 dog, the caudate lobe was able to be visualized during the exploration by using the blunt probe to retract the duodenum. In the other dog, this maneuver was not successful.
Median NASA-TLX for LUS was 9 (IQR, 6.3 to 20) on a scale of 0 (easiest) to 100 (most difficult). There was no significant difference in median NASA-TLX score between dogs weighing < 20 kg (median, 7.5; IQR, 5.7 to 8.7) and those weighing 20 kg and above (median, 15.7; IQR, 9.3 to 25.3; P = .11). There was a strong negative correlation between NASA-TLX score and trial number (r = −0.84; P = .0006).
Elective procedures
Elective procedures performed included laparoscopic ovariectomy (n = 6), laparoscopic gastropexy (4), or concurrent laparoscopic ovariectomy and gastropexy (2). Additional ports were placed in 6 dogs, all of which had a laparoscopic gastropexy performed, to facilitate intracorporeal suturing. In 4 dogs, 2 additional 5-mm ports were placed on the midline cranial to the SILS port, and in 2 dogs, only 1 additional 5-mm port was placed on the midline cranial to the SILS port. All additional ports were placed after laparoscopic exploration and LUS for the purpose of the elective procedure.
Dogs were discharged from the hospital a median of 1 day postoperatively (range, 0 to 1 days). Nine out of 12 (75%) dogs returned 2 weeks postoperatively for an incisional recheck: all dogs’ incision site(s) were completely healed with no incisional complications. Three dogs did not return for recheck evaluation: owners or primary veterinarians of these owners were contacted, and all incisions were reportedly completely healed with no complications noted.
Discussion
In this study, LUS of the liver was feasible and could be efficiently and safely performed using a single incision in all dogs ranging in weight from 5.7 to 62 kg. Access to the right lateral and caudate lobes of the liver was the most limited and challenging, and access to the right division of the liver may not be possible in all dogs. However, dog positioning with a tilt table as well as the use of a laparoscopic blunt probe facilitated accessing these lobes in most dogs. Additionally, over time the ease of performing LUS improved and time to perform LUS decreased, indicating that there is a learning curve initially experienced with this procedure.
There are multiple indications for laparoscopic evaluation and biopsy of the liver including chronically elevated liver enzymes, determining etiology of liver lesions discovered on US or computed CT, and staging of hepatic neoplasia. Prior studies6,18 have demonstrated that laparoscopic liver biopsies are not only safe with low morbidity and mortality rates but also provide sufficient samples for histopathologic evaluation. While US-guided fine needle aspiration of lesions is less invasive, results of cytology tend to be less accurate and have poor agreement with histopathologic findings.19 Laparoscopic liver biopsy is therefore advantageous, as it offers a minimally invasive option for obtaining a diagnostic sample.
In human medicine, LUS of the liver is frequently performed and is considered the standard of care for accurate staging of hepatic neoplasia before definitive surgical treatment.20 LUS provides valuable staging information and has been shown to be more sensitive when detecting small lesions < 10 mm when compared to contrast-enhanced CT.20–22 Additionally, LUS is used to guide laparoscopic surgical procedures including microwave and radiofrequency ablation, as well as laparoscopic liver resections.1,2,10,23 Therefore, LUS may also be valuable in dogs when undergoing a laparoscopic liver biopsy, not only for staging purposes but also for the guidance of liver biopsy to ensure a diagnostic sample is achieved. This study demonstrates that laparoscopic ultrasonographic evaluation of all liver lobes is not only feasible in dogs but can be performed with minimal addition to anesthetic and surgical time.
Minor complications did occur in 2 dogs during the laparoscopic exploration. In both dogs, the laparoscopic blunt probe caused a laceration in a liver lobe when an attempt was made to manipulate or retract the lobe to allow complete visualization. This was likely due to an excessive amount of pressure or force due to a lack of tactile feedback associated with laparoscopic surgery. Therefore, the authors caution that care should be taken when manipulating lobes of the liver or other organs that may be obscuring the liver. Additionally, the use of alternative surgical instruments, such as fan retractors, may aid in preventing complications. However, in both dogs, no medical or surgical intervention was required, and they recovered uneventfully.
Several challenges were experienced when performing LUS. First, access to the caudate and right lateral lobes proved to be the most challenging. In most dogs, the right lateral lobe of the liver was displaced cranially, which prevented visualization of the parietal surface and, thus, prevented placement of the LUS probe on the parenchymal surface. Additionally, in some dogs, the caudate lobe was obscured by the gastrointestinal tract. In most dogs, this could be overcome by positioning the dog in reverse Trendelenburg and tilting them to the left, while also using the blunt laparoscopic probe to either separate the right lateral lobe from the body wall or to retract the intestinal tract. However, as demonstrated by our results, this may not be sufficient in all dogs. One of the drawbacks of a single incision laparoscopic surgery port is the clashing of the instruments, which may have made simultaneous retraction and scanning with the LUS probe more difficult. Therefore, it is possible that placement of an additional paramedian port could further improve access to the right division of the liver, as has been previously reported.3,4,6,18
Another challenge encountered was achieving adequate contact with the LUS probe on the apex of some liver lobes. Because the parenchyma of the liver was much narrower in these regions, pressure from the probe resulted in the displacement of the tissue and a limited ultrasonographic image. Therefore, imaging of the apex of the liver lobes may not be complete using LUS. The use of liver-specific retractors may be helpful to address this issue.
Further, US images in both a longitudinal and transverse plane for all liver lobes were not achievable. In some situations, the LUS probe could be articulated enough to allow the acquisition of the transverse plane as well as the longitudinal plane. The specific ultrasound transducer used in this study had a tip that could be articulated 180° in both the sagittal and frontal planes, allowing some flexibility for the orientation of the probe despite a fixed entry site. Again, it is possible that placement of either a left or right paramedian port could facilitate acquisition of US images in the transverse plane by allowing an alternate position of the probe. Placement of additional paramedian ports was not feasible in the current study as these dogs were healthy and undergoing elective procedures unrelated to their liver but may be considered in dogs with liver disease if acquisition of both transverse and longitudinal planes is desired. It is the authors’ opinion that LUS should not be used to replace preoperative imaging modalities such as transabdominal US or CT but instead should be interpreted in conjunction with these modalities and used to guide laparoscopic surgical procedures.
Several limitations of this study exist. Our sample size was relatively small and may have resulted in a type 1 error when evaluating the trend in NASA-TLX and time to perform LUS with each subsequent study trial. All dogs in this study were healthy with no known liver disease or lesions: the presence of abnormalities within the liver may have increased the amount of time to image the liver, as one may spend more time evaluating lesions when discovered. Additionally, all procedures were performed by 1 surgeon with extensive experience in laparoscopic surgery. Therefore, these results may not be applicable to all surgeons, and others may experience more difficulty initially when performing LUS. However, this design was chosen intentionally to allow interpretation of both technical demand and time to perform LUS over time, which would not have been possible with multiple surgeons and this sample size. Finally, the results of this study are subject to bias as the NASA-TLX was a subjective assessment completed by the surgeon participating in the research.
In conclusion, LUS of the liver in dogs is feasible and safe, adding minimal anesthetic time. All liver lobes could be accessed and scanned through a single incision in this study. Technical demand improved over time, indicating that there is a learning curve associated with LUS. Further evaluation of LUS of the liver in clinical dogs is warranted.
Acknowledgments
The authors thank Olivia Lendoiro from the University of Florida for assisting with data collection.
Disclosures
The authors have nothing to disclose. No AI-assisted technologies were used in the generation of this manuscript.
Funding
This study was funded by the generous contribution of the DeBartolo Minimally Invasive Surgery fund at the University of Florida.
References
- 1.↑
Ferrero A, Lo Tesoriere R, Russolillo N. Ultrasound liver map technique for laparoscopic liver resections. World J Surg. 2019;43(10):2607–2611. doi:10.1007/s00268-019-05046-3
- 2.↑
Ferrero A, Lo Tesoriere R, Russolillo N, Viganò L, Forchino F, Capussotti L. Ultrasound-guided laparoscopic liver resections. Surg Endosc. 2015;29(4):1002–1005. doi:10.1007/s00464-014-3762-9
- 3.↑
Spinella G, Valentini S, Spadari A, Fedrigo M. Laparoscopic ultrasonography in six dogs. Vet Radiol Ultrasound. 2006;47(3):283–286. doi:10.1111/j.1740-8261.2006.00141.x
- 4.↑
Michael AE, Case JB, Massari F, et al. Feasibility of laparoscopic liver lobectomy in dogs. Vet Surg. 2021;50(S1):O89–O98. doi:10.1111/vsu.13566
- 5.↑
Oramas A, Case JB, Toskich BB, et al. Laparoscopic access to the liver and application of laparoscopic microwave ablation in 2 dogs with liver neoplasia. Vet Surg. 2019;48(S1):O91–O98. doi:10.1111/vsu.13153
- 6.↑
McDevitt HL, Mayhew PD, Giuffrida MA, Brown DC, Culp WTN, Runge JJ. Short-term clinical outcome of laparoscopic liver biopsy in dogs: 106 cases (2003–2013). J Am Vet Med Assoc. 2016;248(1):83–90. doi:10.2460/javma.248.1.83
- 7.
Hancock RB, Lanz OI, Waldron DR, Duncan RB, Broadstone RV, Hendrix PK. Comparison of postoperative pain after ovariohysterectomy by harmonic scalpel-assisted laparoscopy compared with median celiotomy and ligation in dogs. Vet Surg. 2005;34(3):273–282. doi:10.1111/j.1532-950x.2005.00041.x
- 8.↑
Devitt CM, Cox RE, Hailey JJ. Duration, complications, stress, and pain of open ovariohysterectomy versus a simple method of laparoscopic-assisted ovariohysterectomy in dogs. J Am Vet Med Assoc. 2005;227(6):921–927. doi:10.2460/javma.2005.227.921
- 9.↑
Simo KA, Sereika SE, Newton KN, Gerber DA. Laparoscopic-assisted microwave ablation for hepatocellular carcinoma: safety and efficacy in comparison with radiofrequency ablation. J Surg Oncol. 2011;104(7):822–829. doi:10.1002/jso.21933
- 10.↑
Liu X, Plishker W, Kane TD, et al. Preclinical evaluation of ultrasound-augmented needle navigation for laparoscopic liver ablation. Int J Comput Assist Radiol Surg. 2020;15(5):803–810. doi:10.1007/s11548-020-02164-5
- 11.↑
Dindo D, Demartines N, Clavien PA. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg. 2004;240(2):205–213. doi:10.1097/01.sla.0000133083.54934.ae
- 12.↑
Clavien PA, Barkun J, de Oliveira ML, et al. The Clavien-Dindo classification of surgical complications: five-year experience. Ann Surg. 2009;250(2):187–196. doi:10.1097/SLA.0b013e3181b13ca2
- 13.↑
Hart SG. Nasa-Task Load Index (NASA-TLX); 20 years later. Hum Factors. 2006;50(9):904–908. doi:10.1177/154193120605000909
- 14.↑
Park LS, Pan F, Steffens D, Young J, Hong J. Are surgeons working smarter or harder? a systematic review comparing the physical and mental demands of robotic and laparoscopic or open surgery. World J Surg. 2021;45(7):2066–2080. doi:10.1007/s00268-021-06055-x
- 15.
Balsa IM, Giuffrida MA, Mayhew PD. A randomized controlled trial of three-dimensional versus two-dimensional imaging system on duration of surgery and mental workload for laparoscopic gastropexies in dogs. Vet Surg. 2021;50(5):944–953. doi:10.1111/vsu.13637
- 16.
Bell SW, Kong JCH, Clark DA, et al. The National Aeronautics and Space Administration-task load index: NASA-TLX: evaluation of its use in surgery. ANZ J Surg. 2022;92(11):3022–3028. doi:10.1111/ans.17830
- 17.↑
Ruiz-Rabelo JF, Navarro-Rodriguez E, Di-Stasi LL, et al. Validation of the NASA-TLX score in ongoing assessment of mental workload during a laparoscopic learning curve in bariatric surgery. Obes Surg. 2015;25(12):2451–2456. doi:10.1007/s11695-015-1922-1
- 18.↑
Petre SL, McClaran JK, Bergman PJ, Monette S. Safety and efficacy of laparoscopic hepatic biopsy in dogs: 80 cases (2004–2009). J Am Vet Med Assoc. 2012;240(2):181–185.
- 19.↑
Wang KY, Panciera DL, Al-Rukibat RK, Radi ZA. Accuracy of ultrasound-guided fine-needle aspiration of the liver and cytologic findings in dogs and cats: 97 cases (1990–2000). J Am Vet Med Assoc. 2004;224(1):75–78. doi:10.2460/javma.2004.224.75
- 20.↑
Montorsi M, Santambrogio R, Bianchi P, et al. Laparoscopy with laparoscopic ultrasound for pretreatment staging of hepatocellular carcinoma: a prospective study. J Gastrointest Surg. 2001;5(3):312–315.
- 21.
Santambrogio R, Opocher E, Ceretti AP, et al. Impact of intraoperative ultrasonography in laparoscopic liver surgery. Surg Endosc. 2007;21(2):181–188. doi:10.1007/s00464-005-0738-9
- 22.↑
John TG, Greig JD, Crosbie JL, Miles A, Garden OJ. Superior staging of liver tumors with laparoscopy and laparoscopic ultrasound. Ann Surg. 1994;220(6):711–719. doi:10.1097/00000658-199412000-00002
- 23.↑
Santambrogio R, Costa M, Barabino M, Opocher E. Laparoscopic radiofrequency of hepatocellular carcinoma using ultrasound-guided selective intrahepatic vascular occlusion. Surg Endosc. 2008;22(9):2051–2055. doi:10.1007/s00464-008-9751-0