Authors’ Response

We appreciate Dr Van Kruiningen’s attention to our review article, ³ his comments regarding the characterization of crypt abscesses, and the opportunity to present additional images that reflect the full spectrum of crypt and mucosal changes noted in the in the FVB.129P2-Abcb1a^tm1Bor (Mdr1a–/– Taconic Biosciences, Hudson, NY) model (originally reported in Lencioni et al ⁴ ).

We agree that the crypt figure initially published in 2008 is not a “bona fide” crypt abscess, as defined in human gastrointestinal pathology and noted by Van Kruiningen. The 2008 figure (see Figure 1 here) was published in a non-pathology-focused journal and was chosen for its orientation so that the Comparative Medicine readership would readily appreciate the general histologic changes compared to the normal tissue presented. We regret the imprecise use of “crypt abscess” in the concise figure legend, rather than “a cystically dilated crypt with attenuated epithelium, containing necrotic epithelial cells, fragments of pyknotic nuclei, and strands of mucus,” as suggested by Van Kruiningen. As seen in Figure 2 and in Figures 4 to 6, there were numerous altered crypts we could have used at the expense of optimal orientation. These presentation decisions were repeated when the image was republished in 2016 without amending the original legend. ³ It is notable that the use of “crypt abscess” in the legend for this particular crypt was published in 2 peer-reviewed journals. Therefore, we believe it is worth expanding on Van Kruiningen’s letter to briefly discuss the diagnostic criteria and significance of crypt abscesses, illustrate their morphology in the human, and present the full spectrum of the crypt morphology in the Mdr1a–/– mouse model coinfected with MNV4 and Helicobacter bilis.

Figures 1–2. MNV4 and Helicobacter bilis coinfection, colon, Mdr1a–/– mouse. Figure 1. This is the original image published in Lencioni et al ⁴ and Hsu et al. ³ Hematoxylin and eosin (HE). Figure 2. Two dilated crypts in the same animal as Figure 1 that are lined by attenuated epithelium, one with abundant intraluminal neutrophils (crypt abscess); the other is dilated and contains sloughed and degenerate cells admixed with debris. Note the mild basal accumulation of plasma cells (plasmacytosis). HE. Figure 3. Chronic crypt destructive colitis: ulcerative colitis, colon, human. Mixed but predominately plasmacellular inflammatory infiltrates are equally dense in the deep and superficial lamina propria. A crypt abscess is present. Crypts are unevenly sized and spaced and irregular and branching. HE. Reprinted with permission Carpenter HA, Talley NJ. The importance of clinicopathological correlation in the diagnosis of inflammatory conditions of the colon: histological patterns with clinical implications. Am J Gastroenterol. 2000;95(4):878–896. Figures 4–6. Chronic-active colitis. Immunohistochemistry for neutrophils. The lesion indicated by the arrow is magnified in Figure 5; the boxed area is magnified in Figure 6. Note the varied population of inflammatory cells within the dilated crypts: dense populations of neutrophils or mixed cells with scattered immunolabeling in crypts with low cellularity.

Crypt abscesses are defined as crypts containing intraluminal neutrophils that often are distended and lined by attenuated epithelium (Figs. 2–6). ^1,2,5 They are nonspecific and seen in active colitis, including inflammatory bowel diseases (most commonly in ulcerative colitis but also in Crohn's disease and indeterminate colitis), and acute infectious colitis, among others. ^1,2,5 The presence of neutrophils in the colon, either within the epithelium (cryptitis) or as crypt abscesses coupled with epithelial damage, indicates an active phase in chronic inflammatory bowel diseases. ^1,2,5 Neutrophils may damage the crypts, leading to rupture and destruction. In all species, cyclic inflammation and crypt damage followed by repair lead to distortion of the crypt architecture as the disease process evolves. ^1,5,7 Therefore, a spectrum of crypt inflammation and architectural changes may be expected in human and veterinary patients and in animal models of chronic-active colitis. ^6,7 Figures 1 to 2 and Figures 4 to 6 illustrate the extent and variability of crypt and mucosal changes and serve to formally document the presence of bona fide crypt abscesses in the Mdr1a–/– coinfection model. For comparison, Figure 3 is a previously published example of a crypt abscess in a human with active-phase ulcerative colitis.

We agree that it is important to use precise descriptions and terminology in comparative pathology and welcome this opportunity to clarify our choice of figure and legend text. We hope the letters and figures presented here aid Veterinary Pathology readers in their appreciation of how imprecision may occur, despite collective efforts; the morphology and significance of crypt abscesses; and the challenges associated with fidelity to human descriptions and terminology in the age of “One Health.”

P. M. Treuting Department of Comparative Medicine, University of Washington School of Medicine, Seattle, WA, USA C. Hsu Department of Comparative Medicine, University of Washington School of Medicine, Seattle, WA, USA