• Open Access

Effect of Endoscopic Forceps on Quality of Duodenal Mucosal Biopsy in Healthy Dogs

Authors


  • The study was supported in part by the Hunting Retriever Club Foundation, Boston Scientific (Natwick, MA) and Olympus America (Center Valley, PA).
  • A research abstract was presented at the 2012 ACVIM Forum, New Orleans, Louisiana, May 30 – June 2

Corresponding author: F. Gaschen, School of Veterinary Medicine, Louisiana State University, Baton Rouge, LA 70808; e-mail: fgaschen@lsu.edu

Abstract

Background

Limited data exist regarding influence of endoscopic forceps on duodenal mucosal biopsy quality and adequacy for histologic examination/assessment in dogs.

Hypothesis/Objectives

Hypothesizing that larger forceps would procure superior specimens, we evaluated effect of 6 disposable forceps on duodenal biopsy weight, depth, crush artifact, and adequacy for histologic examination/assessment.

Animals

Seventeen healthy adult dogs.

Methods

Prospective study. Two operators each obtained 4 duodenal specimens from each dog with each forceps. Lightest sample discarded. One pathologist evaluated blindly other 3 specimens. A total of 612 specimens evaluated. Results analyzed by one-way ANOVA of forceps effects with dog as blocking factor. Posthoc pairwise comparisons examined with Tukey's test when indicated.

Results

Biopsies performed with large capacity forceps heavier (10.56 ± 0.90 and 11.6 ± 0.62 mg (mean ± SD) versus 5.55 ± 0.53 to 8.61 ± 0.49; P < .0001) and adequacy for histologic examination/assessment superior to standard oval and ‘pediatric’ (scores 2.52 ± 0.41 and 2.58 ± 0.37 versus 2.08 ± 0.33 and 2.14 ± 0.29; P < .0001). No statistically significant difference in depth scores. Large capacity forceps with spike associated with less crush artifact than all smaller forceps (scores 1.19 ± 0.16 versus 1.38 ± 0.21 to 1.52 ± 0.21; P < .0001). In same size forceps, presence of spike had no effect on crush artifact and adequacy for histologic examination/assessment (P < .0001).

Conclusions and Clinical Importance

Large capacity forceps are superior, providing higher quality and greater numbers of samples achieving adequacy for histologic examination/assessment. Choice of endoscopic biopsy forceps for duodenal samples influences sample volume and diagnostic utility.

Abbreviations
IACUC

Institutional Animal Care and Use Committee

IBD

inflammatory bowel disease

Gastrointestinal endoscopy is a routine procedure in most referral veterinary hospitals. Endoscopic biopsies constitute an important proportion of the gastrointestinal specimens handled by histopathology laboratories. Endoscopy allows veterinarians to identify, localize, and sample the gastrointestinal mucosa in a safe and rapid fashion to confirm the presence of a benign or malignant process. Histopathologic assessment of adequate tissue samples is required for the diagnosis of common and important diseases such as inflammatory bowel disease (IBD), lymphangiectasia, gastrointestinal lymphoma, and other infiltrative intestinal diseases in dogs. The information obtained from mucosal biopsy is generally critical to planning the care and treatment of dogs.[1] In particular, biopsy quality is important in differentiating IBD from lymphoma in both dogs and cats.[1, 2]

Currently, a very large variety of endoscopic biopsy forceps exists, differing in size, shape (round versus oval), and presence or absence of fenestrations, needles (spikes), and alligator teeth. New variations or models of biopsy forceps become regularly available. Ideal forceps should provide adequate specimens for histologic evaluation, have no mechanical problems, be easy to use, allow no possibility of cross-contamination, and be as inexpensive as possible.[3]

It is well accepted in human medicine that the overall quality of a biopsy specimen has a major impact on adequacy for histologic examination and assessment and also that the nature of the endoscopic biopsy forceps can importantly influence the specimen quality.[4, 5] In veterinary medicine, the total number of biopsy specimens has been shown to correlate with the sensitivity of detecting gastrointestinal diseases in dogs and cats.[2] Limited or no veterinary data have been published on the impact of biopsy forceps type on endoscopic biopsy sample quality or adequacy for histologic examination and assessment in the dog.

The objectives of this study were to evaluate and compare the quality and adequacy for histologic examination and assessment of duodenal samples obtained with different forceps designed for upper gastrointestinal endoscopy in healthy dogs, and to determine if there were any variations in specimen quality associated with different endoscopists. Investigations of endoscopy biopsy forceps performed in humans might not be relevant to veterinary patients, given differences in patient size and organ anatomy. However, because previous human studies revealed that larger cups provided specimens of better quality and adequacy for histologic examination and assessment, we hypothesized that the biopsy forceps with the larger cups will provide the best specimens.[5-7] As for other comparisons (shape, presence of a spike), most studies have not detected meaningful differences between forceps styles.[4, 6, 8] Accordingly, our secondary hypothesis was that there would be no difference in quality or adequacy for histologic examination and assessment between samples collected with alligator and oval shaped forceps either with or without a spike.

Materials and Methods

This was a prospective pathologist-blinded study using various disposable forceps to collect mucosal biopsy samples from the duodenum via upper gastrointestinal endoscopy in 17 medium-sized healthy adult mix breed random source research dogs including 7 males and 10 females. The body weight of these dogs varied between 16.3 and 24.0 kg (mean weight 19.5 kg). The animals were enrolled in an unrelated study protocol and were cared for according to the principles of the Institutional Animal Care and Use Committee (IACUC).

Biopsy Forceps

Six disposable forceps of different cup size (large capacity, standard, pediatric), shape (alligator, oval), design (ordinary, spiked), and make (Boston Scientific, Olympus) were used (Table 1 and Fig 1).

Table 1. Effect of biopsy forceps features on biopsy specimen weight, depth, crush artifact, and histologic assessment adequacy
ShapeNumberSizeWeight (mean in mg ± SD)aDepth (mean score ± SD)aCrush Artifact (mean score ± SD)aHistologic Assessment Adequacy (mean score ± SD)a
  1. MWC, minimum working channel, in mm; Jaw O, jaw opening.

  2. a

    For each category separately (weight, depth, crush artifact, and histologic adequacy), values associated with at least a common letter (A, B, C, or D) are considered nonstatistically different (P > .05).

Alligator large capacity with spike12

MWC 2.8

Jaw O n/a

10.56 ± 0.62[A]2.15 ± 0.29[A]1.19 ± 0.16[A]2.52 ± 0.41[A]
Alligator large capacity23

MWC 2.8

Jaw O n/a

11.60 ± 0.90[B]2.17 ± 0.25[A]1.30 ± 0.21[A,B]2.58 ± 0.37[A]
Alligator standard34

MWC 2.8

Jaw O 7.2

8.61 ± 0.49[C]2.09 ± 0.29[A]1.41 ± 0.21[B,C]2.26 ± 0.37[A,B]
Alligator standard with spike45

MWC 2.8

Jaw O 7.2

8.10 ± 0.62[C]2.06 ± 0.29[A]1.44 ± 0.21[B,C]2.26 ± 0.33[A,B]
Standard oval56

MWC 2.8

Jaw O 7.2

8.57 ± 0.57[C]1.98 ± 0.25[A]1.38 ± 0.21[B,C]2.08 ± 0.33[B]
Alligator pediatric67

MWC 2.8

Jaw O 5.0

5.55 ± 0.53[D]2.04 ± 0.25[A]1.52 ± 0.21[C]2.14 ± 0.29[B]
Figure 1.

Photographs of the different disposable forceps used in the study. #1: Alligator large capacity with spike, #2: Alligator large capacity, #3: Alligator standard, #4: Alligator standard with spike, #5: Standard oval, #6: Alligator pediatric.

Collection of Duodenal Biopsies

Upper gastrointestinal endoscopy was performed on 17 anesthetized healthy dogs (displaying no evidence of gastrointestinal disease) using a flexible video-gastroscope (GIF Olympus 160, working channel diameter: 2.8 mm). Biopsies were taken by 2 operators including an experienced endoscopist (F.G.) and a less experienced internal medicine resident (C.G.). The order of biopsy forceps use was determined in a statistically random fashion before each endoscopic procedure and each forceps was disposed of after being used 24 times. The randomization was performed after a nonalgorithmic technique: drawing papers out of a hat for the choice of the forceps and coin randomization for the choice of the operator. Each operator performed 4 duodenal biopsies with each type of forceps, for each dog, leading to a total of 816 biopsies. All biopsies were performed under the same anesthesia for each dog.

Specimens were taken from normal-appearing duodenal mucosa between the major papilla and the caudal duodenal flexure using the turn-and-suction technique.[9] In short, this involves identification of the target mucosa followed by passage of the forceps through the accessory channel of the endoscope. The forceps cup is then opened and the forceps is withdrawn against the tip of the endoscope. The endoscope and forceps are then directed (“turned”) toward the target and suction is employed (“suction”) as the forceps is gently applied against the targeted mucosa and then closed. This “turn-and-suction” technique effectively captures more tissue within the cups of the forceps.[9]

After withdrawal of the forceps, each duodenal tissue specimen was retrieved from the forceps and unfolded using a hypodermic 25-gauge needle, paying specific attention to not stretch or tear the mucosa. The sample was unfolded directly on the biopsy instrument. Subsequently, within a few seconds, each biopsy specimen was placed separately into individual vials containing 10% buffered formalin.

Weighing of Specimens

Vials were filled with less than 1 mL 10% formaldehyde and capped. Each vial was weighed on a digital scale1 (accuracy ±0.01 mg) in the 12 hours before the procedure. One biopsy specimen was placed in each vial, and every vial was again weighed using the same digital scale1 within 24 hours. The weight of each biopsy specimen was then determined by subtraction. After weighing, the lightest biopsy for each forceps type, operator, and dog was discarded (1 of 4). A total of 612 biopsies were therefore further evaluated in the study.

Processing and Analysis of Histopathology Slides

Each of the 612 specimens was individually placed and properly oriented in a labeled plastic histology cassette containing a sponge. The orientation of each biopsy was determined by examination of each piece of tissue with a magnifying glass when placed in the cassette. The biopsy specimens were initially placed by the operator, mucosal side up, submucosa down, on the sponge.

A total of 612 labeled cassettes placed in a container with 10% formalin were shipped to Texas A&M University College of Veterinary Medicine within 28 days of collection. All biopsy specimens were handled, stored, and processed in a similar fashion. Specimens were processed with increasing concentrations of alcohol and orientated with their mucosal side up. They were then rotated 90 degrees by the technician so each specimen was on its side. In such manner, when the paraffin was added, the orientation was ideal so that all aspects of the biopsy would appear on the slide (from submucosa to epithelium). The applied technique has been previously described in detail.[9]

One board-certified pathologist evaluated each slide for the following characteristics:[2, 8]

  • Depth, scored from 1 to 4 where 1 = very superficial, 2 = mucosa only, 3 = muscularis mucosa present in the section, 4 = submucosa present in the section.
  • Crush artifact, scored from 1 to 3 where 1 = minimum (Fig 2A), 2 = intermediate (Fig 2B), 3 = maximum (Fig 2C).
  • Adequacy for histologic examination and assessment scored from 1 to 4 where 1 = “inadequate,” 2 = “marginal,” 3 = “adequate,” 4 = “superior.” These terms were defined as follows: “Inadequate” was defined as tissue that had only villi or subvillus lamina propria, but not both (Fig 3A). “Marginal” was defined as tissue that had at least 1 villus plus subvillus lamina propria, but did not clearly have the full thickness of the subvillus lamina propria extending to the muscularis mucosa (Fig 3B). “Adequate” was defined as tissue that had at least 3 villi and had subvillus lamina propria that extended to the mucosa-muscularis mucosa border, whether or not it included muscularis mucosa (Fig 3C). “Superior” was defined as tissue that had at least 7 villi and had subvillus lamina propria that extended to the mucosa-muscularis mucosa border, whether or not it included muscularis mucosa (Fig 3D).
Figure 2.

Photomicrographs of biopsy samples of canine duodenum illustrating the crush artifact scoring. Hematoxylin and eosin staining. (A) Score 1: Minimum artifact. (B) Score 2: Intermediate artifact. (C) Score 3: Maximum artifact.

Figure 3.

Photomicrographs of biopsy samples of canine duodenum illustrating the histopathology adequacy assessment scoring. Hematoxylin and eosin staining. (A) Score 1: an “inadequate” tissue sample that only has villus tips. (B) Score 2: a “marginal” tissue sample that has at least 1 villus and subvillus lamina propria, but no clear full thickness of subvillus lamina propria extending to the muscularis mucosa. (C) Score 3: an “adequate” tissue sample that has at least 3 villi and encompasses the entire depth of the intestinal mucosa as seen by subvillus lamina propria, which extends to the mucosa-muscularis mucosa border. (D) Score 4: a “superior” tissue sample that has at least 7 villi and encompasses the entire depth of the intestinal mucosa as seen by subvillus lamina propria, which extends to the mucosa-muscularis mucosa border.

The pathologist was blinded to the endoscopist, type of forceps used, and canine identification.

Statistical Analysis

Because of the absence of studies in the veterinary field, we chose a significant difference of 1 standard deviation with a desired power of 80% and a type I error of 5% to perform our power analysis. According to these calculations, the required number of dogs was 15. Effects in the model included Operator, Forceps, and Dog. All response variables were considered on an ordered-metric scale for the purposes of analysis, so the assumption was made that unit differences between scale points were quantifiable and comparable. Because of nonsignificance, effects attributable to Operator were absorbed (deleted) from the model. A one-way ANOVA was then conducted which included Forceps and Dog as effects. As all Forceps types were randomly assigned to all dogs, a Randomized Complete Block Design was used that assumed no interaction between Forceps and Dog. Posthoc pairwise comparisons were examined with the Tukey's test as appropriate. Data were expressed as mean ± SD. All tests were considered significant at P ≤ .05. A commercial statistical package was used for all statistical analyses.2

Results

There was no statistically significant effect of the operator on the overall quality (including weight [P = .74], depth [P = .94], and crush artifact [P = .79]) and adequacy for histologic examination and assessment (P = .74) of the biopsy specimens.

Specimen weight (Table 1) differed significantly among forceps (P < .0001). Biopsies performed with the 2 large-capacity forceps (#1, 2) were significantly heavier compared with any other forceps (#3, 4, 5, 6). The biopsies collected with the “pediatric” forceps (#6) were significantly lighter in comparison to any other forceps. There was a significant superiority of the large-capacity forceps without spike (#2) when compared with the same forceps with spike (#1) but no statistical difference between the regular alligator with spike (#4) or without spike (#3). Overall the large-capacity forceps (#1, 2) provided significantly heavier specimens in comparison with the other forceps.

The mean depth score (Table 1) was not statistically different among forceps (P = .29).

Crush artifact (Table 1) was significantly different among forceps (P < .0001). The large-capacity spike forceps (#1) was significantly associated with less crush artifact than any of the standard or pediatric size forceps (#3, 4, 5, 6). There was no statistical difference between the 2 large-capacity forceps with or without spike and between the standard alligator forceps with or without spike.

Adequacy for histologic examination and assess-ment (Table 1) differed significantly among forceps (P < .0001). The adequacy for histologic examination and assessment of the large-capacity forceps (#1, 2) with or without spike was significantly superior to the “pediatric” forceps (#6) and standard oval forceps (#5) There was no statistical difference between the 2 large-capacity forceps with or without spike or between the Olympus alligator with or without spike.

Discussion

Results of this study demonstrate the superiority of the large-capacity forceps, underscoring that the size of the biopsy forceps cup has a significant impact on quality (as reflected by weight and crush artifact) and histopathologic adequacy of canine duodenal endoscopic biopsies. Neither the presence of the spike nor the shape of the cup (oval versus alligator) had any statistically significant influence on the adequacy for histologic examination and assessment of the biopsy specimens. Based on these results in healthy dogs, the authors recommend the use of biopsy forceps with the largest biopsy cup possible when collecting duodenal mucosal specimens in dogs.

Many studies comparing the utility of various forceps in obtaining adequate biopsy specimen from human patients have been published. They yielded similar results to our study.[4, 6-8, 10-12] Specifically, they concluded that specimens taken with standard forceps (2.4 mm diameter) were significantly better (in quality and histologic accuracy) than those taken with “pediatric” forceps (1.8 mm diameter).[6, 7] As for other comparisons (shape, presence of a spike), the studies did not detect meaningful differences among forceps styles.[4, 6, 8] Alligator-toothed forceps can provide a higher proportion of good quality specimens when evaluating depth of tissue biopsy.[10] However, studies have demonstrated no significant advantage of alligator forceps when adequacy of histologic assessment was used as a comparison criterion.[4, 6, 10-12] Disagreement exists on the proposed benefit of using biopsy forceps with a spike in humans. Some authors showed that tissue samples obtained with forceps without a spike had significantly greater depth compared with those with a spike[8, 10]; whereas others reported that the presence of a spike resulted in deeper specimens.[4] Despite those contradictory findings, the difference was small and did not generally affect diagnostic assessment if several endoscopic biopsy specimens are provided. In addition, these studies demonstrated no effect on the frequency of crush artifact when forceps with or without a spike were used.[4, 8, 10] Current practice guidelines for human upper gastrointestinal endoscopic biopsies recommend the use of alligator forceps with or without a spike.[4, 6, 8, 11, 12] Additional factors in the choice of biopsy forceps include cost and ease of use.

Although objective studies and recommendations regarding optimal endoscopic biopsy of the human gastrointestinal tract have been published, similar data were not readily available for dogs and cats.[1] It could be more difficult to obtain adequate small intestinal mucosal biopsies in dogs than in other species, because the canine duodenal mucosa is thicker.[13] Body weight of dogs does not influence intestinal wall thickness or relative thickness of its layers. Therefore, the size of the dog should not interfere with the results or recommendations.[13] In veterinary medicine, recent studies looked at both sampling quality and sensitivity of endoscopic biopsy for detecting gastric and duodenal lesions.[2, 14-16] Variability in biopsy quality might result from variability in sample submission technique as well as tissue processing in the laboratory.[15] There is a significant impact of tissue sample quality on the sensitivity of endoscopic biopsy for detection of specific gastric and duodenal lesions.[2] Finally, there is poor agreement in histologic assessments among pathologists in spite of the use of a common histopathologic grading scheme. This is attributed to differences in tissue sectioning, processing, and staining.[14] No evidence-based recommendations were made regarding the use of specific disposable endoscopic biopsy forceps in dogs.[1]

Our study was limited practically by the number and variety of biopsy forceps that could be tested. Indeed, the use of 6 different biopsy forceps led to a total of 48 duodenal biopsies per patient and it was impractical to consider adding more forceps. It could have been useful to test the biopsy forceps with the largest cup available, the jumbo forceps.3 The jumbo forceps provides a significantly higher proportion of adequate biopsy specimens than any other forceps (large capacity forceps included), without being associated with additional complications in humans.[5] The jumbo forceps was not included in our study because it requires using endoscopes with a 3.2 mm biopsy channel, which is larger than the channel available on the gastroscope we used. Endoscopes with such large biopsy channels usually have a wider outside diameter, which can render intubation of the duodenum impossible.

An interesting finding of our study is that larger capacity forceps provided superior adequacy for histologic examination and assessment and weight despite showing no difference in the depth of the specimens. Similar findings have also been reported in the human literature.[4, 11] In comparison to our study, another criterion had been taken into consideration: length/size of the biopsy. The samples taken with the larger capacity forceps showed no difference in depth but significant difference in length and weight.[4, 11] It appeared then that an improved length and weight of the samples without an improved depth could be sufficient to improve adequacy for histologic examination and assessment and ability to provide an appropriate diagnosis.[4, 11] Similarly in our study, biopsy forceps that provided the best adequacy for histologic examination and assessment also collected heavier tissue specimen; however, length data were not recorded.

Overall, our results are congruent with current recommendations in human medicine, but one could argue that different results might be expected in duodenal sampling in healthy cats or in dogs and cats with upper gastrointestinal disease. Indeed, it is the authors' impression that dogs and cats with upper gastrointestinal disease are likely to have a more fragile and friable mucosa. Therefore, the influence of various biopsy forceps shape and size may be different under such circumstances.

In conclusion, the information gathered in this study provides substantial preliminary data that could help in establishing initial guidelines for optimal use of endoscopic biopsy forceps in canine duodenal sampling. On the basis of the results we obtained in healthy dogs, we would recommend the use of large-capacity endoscopic biopsy forceps for duodenal biopsies in dogs. However, further studies evaluating forceps in dogs with gastrointestinal disease should be considered.

Acknowledgments

This study was supported in part by grants from the Hunting Retriever Club Foundation and from the Department of Veterinary Clinical Sciences, School of Veterinary Medicine, Louisiana State University. The evaluated biopsy forceps were made available by Boston Scientific, Natwick, MA and Olympus America, Center Valley, PA. The authors thank Drs M. Willard and R. Stout for their contribution to the study, Mr Kearney for his help with statistical analysis, as well as Ms Roxanne Coffey and Mr Kyle Waite for their excellent technical assistance.

Conflict of Interest Declaration: Authors disclose no conflict of interest.

Footnotes

  1. 1

    Genemate model TG-64; ISC Bioexpress, Kaysville, UT

  2. 2

    SAS version 9.2 Carey, NC

  3. 3

    M00513370; Radial Jaw 4 Jumbo Capacity, Boston Scientific, Natick, MA

  4. 4

    M00513330, Alligator Radial Spike Jaw 4 Large Capacity, Boston Scientific

  5. 5

    M00513320, Alligator Radial Jaw 4 Large Capacity, Boston Scientific

  6. 6

    FB-210K.B, Olympus America, Center Valley, PA

  7. 7

    FB-220K.B, Olympus America

  8. 8

    FB-230K.B, Olympus America

  9. 9

    FB-211K.B, Olympus America

Ancillary