Intraoperative ultrasoundROEN
7.5 MHz · 7.2 cm

Click or drag on the liver: move the probe · right-button drag or the arrows bottom right: turn the liver

Loading the liver (3.4 MB)

Laparoscopic hepato-pancreato-biliary surgery

Laparoscopic ultrasound on a real patient’s liver: find the tumour, stain the portal territory, plan the cut

In laparoscopic liver surgery, as a rule, nobody touches the liver with a hand. A 2.0-centimetre metastasis, and the veins beside it, are found with an ultrasound probe placed on the surface. Here you hold that probe, on the anatomy of one of 197 patients operated on at Memorial Sloan Kettering, New York, whose CT scans, vessel outlines and liver remnants after surgery, as reconstructed by the authors, were published as open data 1.

Drag the probe across the liver ↓

What the hand can no longer feel

In open surgery the surgeon holds the liver and feels a hard one-centimetre nodule between the fingers. In laparoscopy the hand stays outside, and the camera sees only Glisson’s capsule, smooth and glistening.

This liver measures 1,790 mL. It hides a colorectal metastasis of 4.1 mL, about 2.0 cm across, 1.1 mm from a hepatic vein and touching a portal branch at the resolution of the CT. Nothing shows at the surface. The liver turns transparent only on screen: this is how its vessels and the tumour look on the preoperative CT.

  • portal branches
  • hepatic veins and cava
  • the metastasis

A slice beneath the probe

The laparoscopic probe is a linear transducer on a flexible tip, working at 5–10 MHz and seeing 4–10 cm deep 11; the Southampton consensus considers intraoperative ultrasound essential in laparoscopic liver surgery 12. Here the array is 4 cm long. The image is a slice of tissue perpendicular to the surface it touches: the top is the capsule under the probe, the bottom is depth, here 7.2 cm. The yellow mark on the probe head shows which end is the left of the image.

In the transparent view the slice appears where it lies, inside the liver. Move the probe by clicking or dragging on the surface; hold the right button and drag to turn the liver (two fingers on a phone). Rotate the probe with the ⟲ ⟳ buttons or the Q and E keys, and tilt it with “Probe tilt”.

The image is computed as you move, from this patient’s segmentation: an echogenicity for each tissue, attenuation along each scan line, speckle fixed in the tissue. The image is computed by a model.

Where the veins meet the cava

Below the diaphragm the hepatic veins drain into the inferior vena cava. In colour Doppler, red and blue mean “towards the probe” and “away from it”, whether the vessel is an artery or a vein: tilt the probe and the same vein can change colour. Where a vessel runs across the beam, the colour disappears.

Beneath the image, the spectrum from the Doppler gate shows hepatic-vein flow, conventionally called triphasic: two waves towards the heart and a brief reversed wave, all set by the right heart 10. Tap a vessel on the image to move the gate. The velocities are typical values chosen for the model.

Portal pedicle or hepatic vein?

Two kinds of black tube cross the liver. Portal branches run inside Glisson’s sheath with the artery and the bile duct, so their wall is thick and bright. Hepatic veins have almost no visible wall.

Tap the image on what you think is a portal branch.

The metastasis, against the vein

There it is: a rounded lesion with a dark rim, the target pattern of many colorectal metastases, next to a hepatic vein. On the preoperative CT the gap between them is 1.1 mm.

This is where intraoperative ultrasound decides the operation: follow the vein and detach the tumour from it, or take the vein with the liver it drains. In an observational cohort in open surgery, after the tumour was detached from a major vessel the recurrence rate at the cut surface was 4.3% per patient, not significantly different from 5.3% after R0 resection, compared with 19.6% when tumour cells reached a cut through parenchyma 8. Torzilli’s team operated on 93 patients with this ultrasound-guided strategy and saw no recurrence at the cut surface over a mean follow-up of 18 months 7.

Examining the whole liver

Preoperative imaging misses lesions. In 100 patients examined after helical CT, intraoperative ultrasound found more metastases in 23 5; with contrast, ultrasound found 14 further lesions in 6 of 127 patients 6. The probe sees only the slice beneath it, so the liver is examined in overlapping sweeps.

Sweep this liver. The parts your beam has crossed turn blue, and the counter shows how much of the liver you have examined, to a depth of 7 cm.

Staining a portal territory

In 1985 Makuuchi, Hasegawa and Yamazaki described a resection guided by ultrasound: a needle placed under ultrasound control and a dye injected, so the subsegment to be removed shows on the surface 3. With indocyanine green injected into the portal vein, the territory glows in near-infrared light 4.

Adjust the angle and depth until the needle tip reaches the portal branch on the image, then inject. The stained volume is computed from this liver’s portal tree: each point of the liver belongs to the nearest portal branch visible on CT. It is an approximation: real portal territories do not follow the classic planes exactly 13.

Draw the transection line

Click the liver surface to place diathermy marks, at least two; right-button drag turns the liver. The transection plane runs into the liver along your line of sight.

The panel computes for your cut: the closest distance to the tumour, the liver that remains, the vessels you would divide, and the remaining liver that, in this model, loses its portal inflow or its venous drainage. Below it, this patient’s operation as reconstructed by the authors.

Margin width matters less than one might think: in 557 patients a clear margin under 1 cm did not shorten survival 9. The percentage shown is the volume left of the measured liver, tumour included. The remnant thresholds used in practice refer to total liver volume: about 20% for a healthy liver, 30% after intensive chemotherapy, 40% in compensated cirrhosis; below them, portal vein embolisation is discussed 14.

The probe is yours

Every mode at once: transparent liver, the ultrasound slice in place, Doppler, labels, fluorescence, coverage and the transection line. The buttons below switch them on and off.

197 operations, one plate

Each shape is one of the 197 livers, seen from the front at the same scale and sorted by how much liver was left. The pale part was removed; gold marks tumour. The outlined shape is the liver in the simulator. The authors’ selection excluded ablation and patients with more than three wedge resections; the resections range from a few tens of millilitres to more than three quarters of the liver 1.

Half the patients kept less than 67% of their liver; 32 were left with under 30%. On scans with slices 2.5 mm apart or less, 45 of 94 patients had a tumour within 1 mm, one or two voxels, of a portal branch or hepatic vein: the situation in which ultrasound in theatre helps decide whether the vessel can be kept.

How it is made, and what it is not

The data. The Colorectal-Liver-Metastases collection of The Cancer Imaging Archive (CC BY 4.0 licence) holds, for 197 patients operated on for colorectal liver metastases at Memorial Sloan Kettering Cancer Center, the portal-venous CT taken no more than six weeks before surgery and outlines of the liver, the portal and hepatic veins and each tumour 2. It also holds the liver remnant after the resection performed, reconstructed by the authors on the 3D model from postoperative imaging and the margin width at pathology 1.

The liver in the simulator is patient CRLM-CT-1107’s, chosen because the scan has slices 1.5 mm apart and a single metastasis next to a hepatic vein. The outlines were turned into distance fields on a 1.25 mm grid; the vena cava, not outlined in the collection, was tracked slice by slice on the CT from the confluence of the hepatic veins.

The ultrasound is a model: each tissue gets an echogenicity and an attenuation (parenchyma, Glisson’s sheath, vessel lumen, tumour halo and core; outside the liver, by CT density: fat, air, bone), attenuation accumulates along each scan line and is compensated by depth gain, and speckle comes from noise fixed in the tissue. Doppler flow follows the direction of the vascular tree (towards the periphery in the portal vein, towards the heart in the hepatic veins and cava); its sign comes from projection on the beam, and the velocities and waveform are typical values, not measurements.

Territories and transection. Each point of the liver is assigned to the nearest visible portal (and hepatic) branch; branches too small to show on CT are missing. Liver “without inflow” or “without drainage” is the remaining part whose branches, in this tree, connect to the root only through the removed part. The margin is the geometric distance between your plane and the tumour on CT.

What it is not. It is not a planning tool, it has not been validated against real ultrasound, and it does not replace examining a patient. The atlas numbers describe 197 operations selected by the authors’ criteria (no ablation, no more than three wedge resections) and cannot be generalised. CT slice spacing ranges from 0.8 to 7.5 mm; distances along the body axis are less precise where slices are far apart.

Sources

  1. Simpson AL, Peoples J, Creasy JM, et al. Preoperative CT and survival data for patients undergoing resection of colorectal liver metastases. Sci Data 2024;11:172. doi.org/10.1038/s41597-024-02981-2
  2. Colorectal-Liver-Metastases (TCIA), version 2, 2023. CC BY 4.0. doi.org/10.7937/QXK2-QG03
  3. Makuuchi M, Hasegawa H, Yamazaki S. Ultrasonically guided subsegmentectomy. Surg Gynecol Obstet 1985;161:346–350. pubmed.ncbi.nlm.nih.gov/2996162/
  4. Aoki T, Yasuda D, Shimizu Y, et al. Image-guided liver mapping using fluorescence navigation system with indocyanine green for anatomical hepatic resection. World J Surg 2008;32:1763–1767. doi.org/10.1007/s00268-008-9620-y
  5. Sietses C, Meijerink MR, Meijer S, van den Tol MP. The impact of intraoperative ultrasonography on the surgical treatment of patients with colorectal liver metastases. Surg Endosc 2010;24:1917–1922. doi.org/10.1007/s00464-009-0874-8
  6. Torzilli G, Botea F, Donadon M, et al. Criteria for the selective use of contrast-enhanced intra-operative ultrasound during surgery for colorectal liver metastases. HPB 2014;16:994–1001. doi.org/10.1111/hpb.12272
  7. Torzilli G, Montorsi M, Donadon M, et al. “Radical but conservative” is the main goal for ultrasonography-guided liver resection: prospective validation of this approach. J Am Coll Surg 2005;201:517–528. doi.org/10.1016/j.jamcollsurg.2005.04.026
  8. Viganò L, Procopio F, Cimino MM, et al. Is tumor detachment from vascular structures equivalent to R0 resection in surgery for colorectal liver metastases? An observational cohort. Ann Surg Oncol 2016;23:1352–1360. doi.org/10.1245/s10434-015-5009-y
  9. Pawlik TM, Scoggins CR, Zorzi D, et al. Effect of surgical margin status on survival and site of recurrence after hepatic resection for colorectal metastases. Ann Surg 2005;241:715–722. doi.org/10.1097/01.sla.0000160703.75808.7d
  10. McNaughton DA, Abu-Yousef MM. Doppler US of the liver made simple. RadioGraphics 2011;31:161–188. doi.org/10.1148/rg.311105093
  11. SAGES. Guidelines for the use of laparoscopic ultrasound. www.sages.org/publications/guidelines/guidelines-for-the-use-of-laparoscopic-ultrasound/
  12. Abu Hilal M, Aldrighetti L, Dagher I, et al. The Southampton consensus guidelines for laparoscopic liver surgery. Ann Surg 2018;268:11–18. doi.org/10.1097/SLA.0000000000002524
  13. Fasel JHD, Selle D, Evertsz CJ, et al. Segmental anatomy of the liver: poor correlation with CT. Radiology 1998;206:151–156. doi.org/10.1148/radiology.206.1.9423665
  14. Vauthey JN, Dixon E, Abdalla EK, et al. Pretreatment assessment of hepatocellular carcinoma: expert consensus statement. HPB 2010;12:289–299. doi.org/10.1111/j.1477-2574.2010.00181.x