Laparoscopic liver biopsies offer a minimally invasive approach to procuring diagnostic samples of the liver. Other minimally invasive biopsy methods (ie, percutaneous needle biopsies) yield a smaller sample, do not allow visual inspection of the liver to aid in sample selection, and may have bleeding complications that can be challenging to address.1–5 Although the benefits of laparoscopic biopsies are clear,6–8 different techniques are reported with the same surgical instrument, and the quality of liver biopsy samples from these technique variations has not been investigated. Laparoscopic cup biopsy forceps are one of the most used instruments for laparoscopic liver biopsies. This instrument is cup-shaped, with a small spike on the inside of the jaws to help retain tissue. Although the forceps are available in 3- and 5-mm versions, the larger size is more commonly used to obtain liver samples.4 Previous reports1–3 have compared this instrument to other instruments (pretied loops, biopsy needles, punch biopsies), including the gold standard of open surgical wedge biopsy. Biopsy samples yielding adequate portal triads for an accurate diagnosis can be procured with all the tested instruments, but problems with tearing and fragmentation artifacts persist.1 Previous work1–4 reported results using the laparoscopic cup forceps in various ways compared to other instruments, but never investigated the differences in artifact creation among the cup biopsy techniques.
Artifacts reduce the available tissue for evaluation, leading to increased inaccuracies and the potential for misdiagnosis.9,10 Concerns surrounding the effects of laparoscopic techniques on diagnostic accuracy resulting from artifact creation exist, and specific strategies to decrease these artifacts have been recommended. One proposed approach to mitigate fragmentation artifacts requires pushing the laparoscopic cannula down to the surface of the liver (or the liver to be pulled to the cannula) to pull the liver sample encapsulated within the forceps cup gently out of the patient.1 Unfortunately, this technique cannot always be performed, depending on the conformation of the patient, placement of the ports, the region of interest in the liver, and consistency of the liver parenchyma (friable vs fibrous). In our experience, if mitigation techniques are not possible, the decision is sometimes made to move forward with an open surgical approach instead of risking inaccurate samples or foregoing sampling entirely.
The 3 laparoscopic cup biopsy techniques used frequently in practice include the pull, twist, and cannula approaches. Each biopsy technique has advantages and disadvantages regarding how the biopsy can be performed and what samples can be procured. However, all have the benefit of decreased pain compared to an open approach, and visualization of the biopsy site compared to closed needle techniques. The objective of our study was to determine the difference in histologic artifacts and morphologic diagnosis among 3 laparoscopic cup biopsy forceps techniques and wedge hepatic samples. Our primary hypothesis was that the cannula technique would not create fewer artifacts compared to other laparoscopic techniques. Our secondary hypothesis was that there would not be a difference between these laparoscopic techniques and the gold standard wedge biopsy regarding diagnosis.
Materials and Methods
Animals
Client-owned dogs presented to Cornell University Hospital for Animals emergency service for euthanasia between January 3 and July 29, 2021, for reasons unrelated to our study, were eligible for enrollment. The study was awarded an exemption by the Institutional Animal Care and Use Committee at Cornell University (Protocol ID No. 031021-09). Animals were included in the study if owners consented to inclusion, and samples could be collected within 3 hours of euthanasia.3 Animals were not excluded if they had known hepatic disease, as long as adequate samples could be obtained to compare histologic results among the techniques. For this reason, 3 liver lobes were sampled from each cadaver to mimic clinical and surgical objectives and procedures. Animals were left at room temperature until samples were collected. Our study was modeled after previous research1–4 comparing laparoscopic liver biopsy techniques to the gold standard wedge biopsy.
Cadavers were placed in dorsal recumbency and the ventral abdomen was prepared by clipping the hair on the midline. A 5-cm incision was made through the skin, subcutaneous tissue, and linea alba on the cranial midline just caudal to the xiphoid. The abdomen was not insufflated because of the inaccessibility of the laparoscopic tower during harvest opportunities. A small wound retractor (SurgiSleeve Wound Protector; Medtronic) was placed in this incision, and 4 different biopsy techniques were performed in 3 different liver lobes. This yielded 3 samples of each biopsy method and 12 samples per dog. Three liver lobes were used to ensure accuracy with histologic diagnosis as implemented in previous a report.4 All biopsies were performed by the same individual (NB) with extensive experience in minimally invasive surgery to reduce any variation in technique. Techniques were performed as they would be in a live patient, including holding the jaws of the biopsy instrument closed for 15 seconds after grasping to decrease hemorrhage. All samples were obtained from the liver margin.
Laparoscopic techniques used 5-mm cup biopsy forceps (Karl Storz) grasping the liver margin, closing the jaws, and holding them together for 15 seconds before pulling the sample free forcefully (the PULL technique; performed first per lobe; Supplementary Video S1), rotating the forceps 360° in 1 direction until the sample was freed (the TWST technique; performed second per lobe; Supplementary Video S2), or pulling the forceps caudally through a 5-mm cannula (Geniport Pyramidal tip trocar and cannula system; Genicon) to remove the sample (the CAN technique; performed third per lobe; Supplementary Video S3). To perform the CAN technique, the biopsy forceps were pulled caudally until the cannula touched the liver’s surface. Caudal tension on the biopsy forceps with cranial movement of the cannula was used to withdraw the biopsy forceps into the cannula, dissecting the liver with the cannula.3 The fourth technique performed per lobe consisted of the use of Metzenbaum scissors to obtain a 1 X 1 X 1-cm wedge biopsy sample that acted as the gold standard control (CON). The CON biopsy sample characteristics (eg, artifacts, portal triads) and diagnosis were considered the gold standard by which all other samples were compared. All liver biopsy samples were removed gently with a 25-gauge needle from the cup and placed in formalin. The abdomen was then closed routinely with 2-0 or 3-0 nylon suture, and the animals were disposed of per the owners’ wishes.
Histopathologic data
Liver biopsy samples were placed into separate cassettes based on sampling location and then immersed in neutral-buffered 10% formalin at room temperature. After 24 to 72 hours of formalin fixation, samples were processed and embedded in paraffin for sectioning. Sections were cut at a 5-µm thickness, mounted on glass slides by the same technician, and stained with H&E stain according to standard procedures. Each dog’s 3 samples of the same technique were sectioned on the same slide for review. The resultant slides were numbered randomly by a technician preparing the slides who retained the key for unblinding after the study. A board-certified veterinary pathologist (ADM) reviewed each slide without knowing which slide came from which dog, the history, or the biopsy technique until the conclusion of data collection.
Fifteen histologic features were evaluated: neoplasia, hepatocellular atrophy, biliary hyperplasia, nodular hyperplasia, nodular regeneration, cholangiohepatitis, copper, iron, lipofuscin, fibrosis, cirrhosis, cholestasis, extramedullary hemopoiesis, necrosis, and thrombosis were recorded as present or absent. Nonneoplastic features were recorded and scored on a scale of 0 to 3, with 3 being the most severe; these included congestion, vacuolar hepatopathy, tissue inflammation, and vascular abnormalities. Definitions for all histologic features have been described previously.10,11 A morphologic diagnosis based on the World Small Animal Veterinary Association (WSAVA) Liver Standardization Group’s classification10 of hepatic disorders was established using the gold standard wedge after the history and overall picture of the case was presented to the pathologist. Sample quality and artifact types were also documented. Artifacts were categorized as tearing or fragmentation, and presence of ragged edges, and were graded subjectively as none present, mild, moderate, or severe. Sample quality was categorized as poor, good, or excellent. Morphologic diagnoses included normal, vacuolar hepatopathy, neoplasia, primary fibrosis, acute hepatitis, chronic hepatitis, congestion, necrosis, cholangiohepatitis, reactive hepatitis, cholestasis, vascular proliferation or anomaly, and peritonitis. Samples were allowed to have > 1 morphologic diagnosis.
Statistical analysis
Statistical analyses were completed using standard commercial software (R version 4.1.1; R Foundation for Statistical Computing) or GraphPad Prism 9 (GraphPad). Gwet AC1 statistics12 and an interclass correlation coefficient (ICC; R version 0841; R Foundation for Statistical Computing) were calculated using standard software. Gwet AC1 unweighted agreement statistics were calculated for variables measured on a nominal scale (these include neoplasia, hepatocellular atrophy, biliary hyperplasia, nodular hyperplasia, vacuolar hepatopathy, cholangiohepatitis, copper, iron, lipofuscin, fibrosis, inflammation, cirrhosis, cholestasis, extramedullary hematopoiesis, necrosis, thrombosis, sample quality, and artifacts); ordinal weights were used for variables measured on an ordinal scale (congestion and vascular abnormalities). An ICC was calculated for weight because it was measured on a continuous scale. In addition, we compared weights among sampling techniques with the Friedman test. Post hoc analysis with the Dunn multiple comparison test was performed with adjusted P values. A value of P < .05 established significance for all tests. We assessed normality with the Kolmogorov-Smirnov test. Gaussian data are reported as mean and SD; non-Gaussian data are reported as median and range.
The Gwet AC1 statistic was used because it is not influenced by prevalence. In contrast, Cohen kappa can be skewed when there is high agreement among raters and a large imbalance in the distribution of the categorical responses. Method agreement was characterized based on the Altman benchmarking scale12 and applied to Gwet AC1 or ICC as poor (< 0.2), fair (0.21 to 0.4) moderate (0.41 to 0.6), excellent (0.61 to 0.8), and very good (> 0.8). A Kruskal-Wallis test was performed to assess for differences among the groups.
Results
Twenty canine cadavers were included in our study. The median age of the dogs at the time of euthanasia was 8.5 years (range, 3 to 13 years), and the median body weight was 30 kg (range, 3.5 to 70 kg). Sex distribution included 1 sexually intact male, 9 castrated males, 1 sexually intact female, and 9 spayed females. The study included 60 liver lobes from 20 dogs (3 liver lobes per dog), with 4 matched samples (PULL, TWST, CAN, and CON) from each sampled lobe, yielding 12 samples per cadaver, 240 total liver samples, and 60 liver samples per technique.
The CON biopsy yielded significantly larger samples (P < .001; mean ± SD, 0.6 ± 0.17 g) than the PULL (mean ± SD, 0.4 ± 0.11 g) or CAN (median, 0.38 g; range, 0.29 to 0.6 g) techniques. The TWST method produced the second-largest samples (mean ± SD, 0.47 ± 0.06 g), which were significantly larger (P = .035) than PULL or CAN, but not significantly smaller than CON.
Of the 240 total liver samples, there were 12 morphologic diagnoses (Table 1), with 45% (108/240) of liver samples being considered histologically normal. Neoplasia was diagnosed in none of the samples. There was excellent agreement among all sampling techniques for most diagnostic features (Gwet AC1, 0.93 to 1), with good agreement found only for congestion (Table 2). Sample quality agreement among techniques was good when comparing the laparoscopic and wedge techniques.
Number and percentages (with 95% CIs) of morphologic diagnoses identified on histologic examination of 240 liver biopsy samples obtained by wedge biopsy (control; n = 60) versus 3 laparoscopic cup biopsy forceps techniques that involved grasping the liver margin with a 5-mm cup forceps for 15 seconds before either forcefully pulling the sample free (the PULL technique; n = 60), twisting the captured sample 360° in 1 direction until the sample was free (the TWST technique; n = 60), or pulling the captured sample until the liver’s surface touched the laparoscopy cannula, which was then advanced over the forceps, freeing the forceps and sample (the CAN technique; n = 60) from cadavers of 20 client-owned dogs euthanized for unrelated reasons between January 3 and July 29, 2021.
| Diagnosis | No. of liver samples (N = 240) | Total liver samples, % (95% CI) |
|---|---|---|
| Histologically normal | 108 | 45 (39–51) |
| Mild to moderate, multifocal chronic vacuolar hepatopathy (degenerative glycogen type) | 3 | 1.3 (0.25–3.8) |
| Mild to moderate, multifocal chronic vacuolar hepatopathy with rare necrosis (degenerative glycogen type) | 9 | 3.8 (1.2–7.0) |
| Moderate, acute, centrilobular necrosis | 12 | 5 (2.8–8.6) |
| Severe, acute, centrilobular necrosis | 12 | 5 (2.8–8.6) |
| Hypoperfusion (mild phenotype) | 24 | 10 (6.8–14.5) |
| Mild, multifocal, chronic vacuolar hepatopathy (degenerative glycogen type) | 12 | 5 (2.8–8.6) |
| Moderate, multifocal, chronic vacuolar hepatopathy (glycogen type) | 15 | 6.3 (3.8–10.1) |
| Moderate, multifocal, chronic vacuolar hepatopathy (degenerative and nondegenerative glycogen type) | 9 | 3.8 (1.2–7.0) |
| Mild to moderate, acute midzonal necrosis with bile sludging and mild vacuolar hepatopathy | 12 | 5 (2.8–8.6) |
| Severe, acute, midzonal, and centrilobular necrosis; moderate, chronic lymphoplasmacytic portal hepatitis; mild canalicular cholestasis | 12 | 5 (2.8–8.6) |
| Mild, diffuse, chronic lymphoplasmacytic portal hepatitis (nonspecific) | 12 | 5 (2.8–8.6) |
Results of analysis to detect agreement (Gwet AC1 unweighted agreement statistic [Gwet AC1] or interclass correlation coefficient) among the 4 biopsy techniques used to obtain the hepatic biopsy samples described in Table 1 on the basis of histologic diagnostic feature.
| Diagnostic feature | Agreement statistic | P value |
|---|---|---|
| Neoplasia | 1 | NaN |
| Hepatocellular atrophy | 1 | NaN |
| Biliary hyperplasia | 1 | < .001 |
| Congestion | 0.65854 | 2.15E-06c |
| Nodular hyperplasia | 1 | < .001 |
| Nodular regeneration | 1 | < .001 |
| Vacuolar hepatopathy | 0.92714 | 2.9E-13c |
| Cholangiohepatitis | 1 | NaN |
| Copper | 1 | < .001 |
| Iron | 0.96714 | < .001 |
| Lipofuscin | 1 | < .001 |
| Fibrosis | 1 | < .001 |
| Inflammation, type and chronic | 0.94585 | 4.4E-16c |
| Cirrhosis | 1 | NaN |
| Vascular abnormalities | 0.97374 | < .001 |
| Cholestasis | 1 | < .001 |
| Extramedullary hematopoiesis | 0.97374 | < .001 |
| Necrosis | 0.94679 | 3.1E-13c |
| Thrombosis | 0.97305 | < .001 |
| Sample quality | 0.64438 | 1.32E-05c |
| Artifacts | 0.44269 | 1.12E-05c |
| Sample weight | 0.18522 | .01 |
Reported as Gwet AC1, for which the agreement is considered poor (Gwet AC1, < 0.2), fair (Gwet AC1, 0.21 to 04), moderate (Gwet AC1, 0.41 to 0.6), excellent (Gwet AC1, 0.61 to 0.8), or very good (Gwet AC1, > 0.8).
Reported as or interclass correlation coefficient (ICC), for which the agreement is considered poor (ICC, < 0.2), moderate (ICC, 0.21 to 04), moderate (ICC, 0.41 to 0.6), excellent (ICC, 0.61 to 0.8), or very good (ICC, > 0.8).
Reported as the exponential notation for powers of 10 (eg, 2.15E-06 means 2.15 X 10−4).
Fair agreement among techniques was observed for artifacts (Gwet AC1, 0.44) when comparing all the laparoscopic techniques with the CON biopsy, as the wedge resulted in significantly fewer (P < .001) artifacts (95% [57/60] of samples had crisp edges and 85% [51/60] had no or mild tearing; Figure 1; Table 3). When evaluating the laparoscopic techniques alone, the TWST technique resulted in the best overall artifact profile after the CON, with 90% of samples (54/60) having crisp edges and 65% (39/60) having no or mild tearing. The agreement was good (ICC, 0.73 for edges and 0.76 for tearing) among all cup biopsy forceps techniques. The Kruskal-Wallis test revealed minimal differences among the 3 laparoscopic biopsy cup techniques regarding tearing and edges. There was a slight advantage of the TWST technique for tearing because there was no significant difference between the TWST and CON techniques (Table 4).
Photomicrographs of 4 representative liver biopsy samples obtained by wedge biopsy (A) versus 3 laparoscopic cup biopsy forceps techniques that involved grasping the liver margin with a 5-mm cup forceps for 15 seconds before either twisting the captured sample 360° in 1 direction until the sample was free (B); pulling the captured sample until the liver’s surface touched the laparoscopy cannula, which hat was then advanced over the forceps, freeing the forceps and sample (C); or pulling the sample free forcefully (D) from cadavers of 20 client-owned dogs euthanized for unrelated reasons between January 3 and July 29, 2021. A—Edges are crisp and no artifacts are present. B—A small focal area of extravasated RBCs and serum exudation is present corresponding to mild artifactual change. C—Multiple areas of extravasated RBCs and areas of tissue tearing corresponding to moderate artifactual change are present. D—Liver tissue is tattered and torn, with multiple artifactual tears corresponding to severe artifactual change. H&E stain; bar = 200 µm.
Citation: American Journal of Veterinary Research 83, 12; 10.2460/ajvr.22.08.0127
Photomicrographs of 4 representative liver biopsy samples obtained by wedge biopsy (A) versus 3 laparoscopic cup biopsy forceps techniques that involved grasping the liver margin with a 5-mm cup forceps for 15 seconds before either twisting the captured sample 360° in 1 direction until the sample was free (B); pulling the captured sample until the liver’s surface touched the laparoscopy cannula, which hat was then advanced over the forceps, freeing the forceps and sample (C); or pulling the sample free forcefully (D) from cadavers of 20 client-owned dogs euthanized for unrelated reasons between January 3 and July 29, 2021. A—Edges are crisp and no artifacts are present. B—A small focal area of extravasated RBCs and serum exudation is present corresponding to mild artifactual change. C—Multiple areas of extravasated RBCs and areas of tissue tearing corresponding to moderate artifactual change are present. D—Liver tissue is tattered and torn, with multiple artifactual tears corresponding to severe artifactual change. H&E stain; bar = 200 µm.
Citation: American Journal of Veterinary Research 83, 12; 10.2460/ajvr.22.08.0127
Photomicrographs of 4 representative liver biopsy samples obtained by wedge biopsy (A) versus 3 laparoscopic cup biopsy forceps techniques that involved grasping the liver margin with a 5-mm cup forceps for 15 seconds before either twisting the captured sample 360° in 1 direction until the sample was free (B); pulling the captured sample until the liver’s surface touched the laparoscopy cannula, which hat was then advanced over the forceps, freeing the forceps and sample (C); or pulling the sample free forcefully (D) from cadavers of 20 client-owned dogs euthanized for unrelated reasons between January 3 and July 29, 2021. A—Edges are crisp and no artifacts are present. B—A small focal area of extravasated RBCs and serum exudation is present corresponding to mild artifactual change. C—Multiple areas of extravasated RBCs and areas of tissue tearing corresponding to moderate artifactual change are present. D—Liver tissue is tattered and torn, with multiple artifactual tears corresponding to severe artifactual change. H&E stain; bar = 200 µm.
Citation: American Journal of Veterinary Research 83, 12; 10.2460/ajvr.22.08.0127
Proportions of the liver biopsy samples described in Table 1 stratified by histologic appearance of edges (crisp, equating to the best sample; good; or ragged, as the least favorable) and the presence of tearing of the tissue (none present, mild, moderate, or severe).
| Appearance of edges | Presence of tearing | ||||||
|---|---|---|---|---|---|---|---|
| Technique | Crisp | Good | Ragged | No | Mild | Moderate | Severe |
| CON | 57/60 (95) | 3/60 (5) | 0/60 (0) | 27/60 (45) | 24/60 (40) | 9/60 (15) | 0/60 (0) |
| CANa | 51/60 (80) | 9/60 (15) | 3/60 (5) | 3/60 (5) | 33/60 (55) | 21/60 (35) | 3/60 (5) |
| TWSTb | 54/60 (90) | 0/60 (0) | 6/60 (10) | 9/60 (15) | 30/60 (50) | 15/60 (25) | 6/60 (10) |
| PULLc | 54/60 (90) | 3/60 (5) | 3/60 (5) | 6/60 (10) | 30/60 (50) | 18/60 (30) | 6/60 (10) |
Results of the Kruskal-Wallis test to compare the difference of tearing artifacts between the control and each of the 3 laparoscopic biopsy techniques described in Table 1, and among the laparoscopic techniques themselves.
| Dunn multiple comparisons test | Mean rank difference | Adjusted P value |
|---|---|---|
| CON vs CANa | –18.95 | .03 |
| CON vs TWSTb | –15.40 | .14 |
| CON vs PULLc | –18.25 | .04 |
| CAN vs TWST | 3.55 | > .99 |
| CAN vs PULL | 0.70 | > .99 |
| TWST vs PULL | –2.85 | > .99 |
Discussion
The laparoscopic cup biopsy forceps is one of the most common instruments used for liver biopsies because of the large sample size removed, ease of technique, and availability of 2 sizes. Laparoscopic liver biopsies allow for safe, accurate diagnosis with a low rate of elective or emergent conversions and a low complication rate.6,7,13 The techniques used with this instrument for laparoscopic liver biopsy have not been investigated previously for artifact creation, but adequate diagnostic results have been established.1–4 Other minimally invasive techniques such as transjugular needle biopsy have also been investigated, and fragmentation artifacts have been described.14,15 In a canine cadaver model, a high rate of capsular perforation discouraged its clinical use in veterinary patients.16 The relevance of artifacts in liver samples lies in their potential to obscure diagnostic features and morphologic diagnoses, thereby affecting treatment for the patient. The artifacts we identified were fragmentation or tearing and edge configuration, which we deemed were the most relevant to the techniques and possible histologic diagnoses.17
As expected, the gold standard wedge biopsy technique (CON) did create the largest sample with the fewest artifacts. Although the other techniques had an increased level of artifacts, the pathologist determined the morphologic diagnosis accurately in each case. The agreement was moderate to good when comparing artifacts across laparoscopic techniques, with a slight advantage to the TWST technique for tearing. This finding was unexpected because the TWST technique appears visually to create the most tissue trauma. The TWST technique has been used in previous studies7 and was advocated during our training as a technique that decreases hemorrhage through vasospasm of parenchymal vessels. However, no specific documentation of this effect could be found in the literature. It is possible that by twisting the surrounding parenchyma, the tissue is allowed to separate more naturally along lobular separations, explaining the decreased fragmentation.
The CAN technique was not found to have a better artifact profile. In fact, this technique had more tearing than both other laparoscopic techniques, leading us to accept our primary hypothesis that the CAN technique does not create fewer artifacts compared to other laparoscopic techniques. The CAN technique also provided the smallest samples by weight, consistent with visual inspections of the tissue. Because the cup is pulled through the cannula with this technique, only the tissue within the jaws is extracted. This dragging may also explain the increased number of ragged edges and more fragmentation seen with this technique. The PULL and TWST techniques commonly remove adjacent tissue, leading to larger samples. These 2 techniques can also be performed in the liver lobe’s center or periphery. Because of the liver’s anatomic location within the rib cage, it would be difficult to position a cannula directly over the center of the liver lobes to perform the CAN technique. In our experience, cannulas are almost always placed caudal to the liver margin. A PULL or TWST technique is the only possible sampling method if a central lesion is visualized. The wedge technique is also not viable for central lesions, making it less valuable for nondiffuse diseases.
There was excellent agreement for all diagnostic features (except congestion) and morphologic diagnosis across the sampling techniques. This result agrees with previous studies3,4 and allows us to accept our secondary hypothesis that there is not a difference between these laparoscopic techniques and the gold standard wedge biopsy regarding diagnosis. This reinforces the widely held opinion that laparoscopic liver biopsies are an accurate technique for diagnosing liver disease compared to an open approach. We followed the recommendations of the WSAVA working group and obtained multiple samples from 3 liver lobes in our study to increase diagnostic accuracy, as is done in clinical patients. Although 1 study3 found the accuracy of all laparoscopic biopsy techniques to be low compared to a larger sample, the results of our study showed excellent agreement among all sampling techniques for peripheral samples. In that study,3 the samples were taken from the center of the liver lobe instead of the periphery, and this may have affected their quality because central biopsies can be more challenging to acquire. A more recent study4 using peripheral samples found that the diagnostic accuracy was comparable between wedge and laparoscopic cup biopsy forceps even when using the smaller 3-mm version.
Laparoscopic liver biopsy is one of the most common procedures performed by minimally invasive clinicians, but the procedure is only appropriate if it yields accurate data for managing patient care. Visualizing the liver parenchyma before sampling and addressing bleeding if necessary are valuable advantages to laparoscopic procedures over percutaneous image-guided techniques. Specific regions or masses (cavitated) can be sampled that might otherwise be considered dangerous.6,7 As concerns persist about the quality of these samples, clinicians are not always willing to recommend this procedure, leaving patients to undergo more invasive surgeries or no surgery at all. Our results support the use of laparoscopic cup biopsy forceps in any of the 3 described methods for acquiring accurate morphologic liver diagnosis. The technique that provided the best overall sample size and artifact profile is one of the techniques that can be used for central or peripheral lesions. This allows clinicians to choose whichever method they feel is safest for sample acquisition instead of imposing specific technique criteria.
Our study had several limitations, including the use of 1 pathologist, the use of cadaver samples, the use of 1 staining technique (H&E stain), and the underrepresentation of certain disease processes in the sampled tissue (neoplasia and cirrhosis). We chose to use 1 pathologist (ADM) with 17 years of veterinary diagnostic pathology experience based on a previous study4 and to ensure uniformity of evaluation, which has been reported3,18 to vary among pathologists. The use of cadaveric tissue was elected to avoid affecting a clinical patient’s diagnosis and removal of excessive tissue; however, this means the samples were not representative of patients with active liver disease, and the effects of euthanasia on hepatic congestion must be considered. Because the results of our study illustrated diagnostic accuracy and acceptable artifact profiles for laparoscopic techniques, a follow-up study in clinical patients is warranted. A live patient study would test for repeatability of these results in diseased liver tissue and may provide insight into the best laparoscopic method for each disease process. In addition, variability among surgeons is another factor worthy of investigation. Live patient studies could also investigate differences among biopsy techniques in blood loss and surgical time. We harvested tissue within 3 hours of euthanasia to decrease the effects of postmortem changes, but changes associated with autolysis could have altered our findings. The effect of many factors such as body weight, breed, and ambient temperature on the degree of autolysis cannot be determined and is a limitation to cadaveric studies. Because liver samples were harvested after hours routinely, the laparoscopic tower was not available, which led us to mimic a laparoscopic-assisted procedure in these cases. We do not feel performing these biopsies under insufflation would affect the histologic artifacts of these samples, but further studies could investigate this concern. Adding stains specific for fibrosis or other hepatic pathology may have led to insights regarding the effect of the underlying disease and artifact production; however, this was outside the scope of our study. Because previous research1–4 has reported obtaining adequate portal triads with this laparoscopic instrument, we did not evaluate this aspect of the liver biopsies and instead focused on morphologic features and artifacts. A previous report19 showed differences between peripheral and central biopsies concerning disease diagnosis. Additional studies could be conducted to evaluate the differences between central and peripheral samples with the laparoscopic cup biopsy forceps and artifact creation. We focused on these techniques because wedge biopsies are always performed at the periphery.
In conclusion, there was no benefit with regard to artifact profile for the CAN technique, and all laparoscopic biopsy techniques identified the morphologic diagnosis appropriately. This supports clinicians’ choice of the technique they feel is most appropriate to access the desired lesion. The TWST technique resulted in the largest laparoscopic sample and had a slight advantage when comparing tearing artifacts, supporting its use for diagnostic procedures of the liver.
Supplementary Materials
Supplementary materials are posted online at the journal website: avmajournals.avma.org
Acknowledgments
No third-party funding or support was received in connection with this study or the writing or publication of the manuscript.
The authors declare that there were no conflicts of interest.
The authors thank Stephen Parry of the Cornell Statistical Consulting Unit for assisting with and supporting the statistical analysis.
References
- 1.↑
Fernandez N, del-Pozo J, Shaw D, Marques AIC. Comparison of two minimally invasive techniques for liver biopsy collection in dogs. J Small Anim Pract. 2017;58(10):555–561. doi:10.1111/jsap.12708
- 2.
Vasanjee SC, Bubenik LJ, Hosgood G, Bauer R. Evaluation of hemorrhage, sample size, and collateral damage for five hepatic biopsy methods in dogs. Vet Surg. 2006;35(1):86–93. doi:10.1111/j.1532-950X.2005.00117.x.
- 3.↑
Kemp SD, Zimmerman KL, Panciera DL, Monroe WE, Leib MS, Lanz OI. A comparison of liver sampling techniques in dogs. J Vet Intern Med. 2015;29(1):51–57. doi:10.1111/jvim.12508.
- 4.↑
Kimbrell TL, Milovancev M, Olsen R, Löhr CV. Comparison of diagnostic accuracy of laparoscopic 3 mm and 5 mm cup biopsies to wedge biopsies of canine livers. J Vet Intern Med. 2018;32(2):701–706. doi:10.1111/jvim.15054.
- 5.↑
Mayhew PD, Weisse C. Liver and biliary system. In: Johnston SA, Tobias K, eds. Veterinary Surgery Small Animal Expert Consult. 2nd ed. Elsevier; 2015:1828–1852.
- 6.↑
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. doi:10.2460/javma.240.2.181.
- 7.↑
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.
- 8.↑
Mayhew PD, Freeman L, Kwan T, Brown DC. Comparison of surgical site infection rates in clean and clean-contaminated wounds in dogs and cats after minimally invasive versus open surgery: 179 cases (2007–2008). J Am Vet Med Assoc. 2012;240(2):193–198. doi:10.2460/javma.240.2.193.
- 9.↑
Rothuizen J, Twedt DC. Liver biopsy techniques. Vet Clin North Am Small Anim Pract. 2009;39(3):469–480. doi:10.1016/j.cvsm.2009.02.006.
- 10.↑
Rothuizen J, Desmet VJ, van den Ingh TSGAM, Twedt DC, Bunch SE, Washabau RJ. Sampling and handling of liver tissue. In: WSAVA Standards for Clinical Histological Diagnosis Canine and Feline Liver Diseases. Saunders Ltd; 2006:5–14.
- 11.↑
Cole TL, Center SA, Flood SN, et al. Diagnostic comparison of needle and wedge biopsy specimens of the liver in dogs and cats. J Am Vet Med Assoc. 2002;220(10):1483–1490. doi:10.2460/javma.2002.220.1483.
- 12.↑
Gwet KL. Handbook of Inter-Rater Reliability: The Definitive Guide to Measuring The Extent Agreement Among Raters. 4th ed. Advanced Analytics LLC; 2014.
- 13.↑
Buote NJ, Kovak-McClaran JR, Schold JD. Conversion from diagnostic laparoscopy to laparotomy: risk factors and occurrence. Vet Surg. 2011;40(1):106–114. doi:10.1111/j.1532-950X.2010.00759.x.
- 14.↑
Cholongitas E, Quaglia A, Samonakis D, et al. Transjugular liver biopsy: how good is it for accurate histological interpretation? Gut. 2006;55(12):1789–1794. doi:10.1136/gut.2005.090415.
- 15.↑
Vibhakorn S, Cholongitas E, Kalambokis G, et al. A comparison of four- versus three-pass transjugular biopsy using a 19-G Tru-Cut needle and a randomized study using a cassette to prevent biopsy fragmentation. Cardiovasc Intervent Radiol. 2009;32(5):508–513. doi:10.1007/s00270-008-9412-7.
- 16.↑
Levien AS, Weisse C, Donovan TA, Berent AC. Assessment of the efficacy and potential complications of transjugular liver biopsy in canine cadavers. J Vet Intern Med. 2014;28(2):338–345. doi:10.1111/jvim.12305.
- 17.↑
Thompson SW, Luna LG. An Atlas of Artifacts Encountered in the Preparation of Microscopic Tissue Sections. Charles C Thomas Publisher; 1978.
- 18.↑
Theodossi A, Skene AM, Portmann B, et al. Observer variation in assessment of liver biopsies including analysis by kappa statistics. Gastroenterology. 1980:79(2):232–241. doi:10.1016/0016-5085(80)90135-3
- 19.↑
Petrelli M, Scheuer PJ. Variation in subcapsular liver structure and its significance in the interpretation of wedge biopsies. J Clin Pathol. 1967;20(5):743–748. doi:10.1136/jcp.20.5.743
