Know variables in gastrointestinal tissue histopathology

Clinicians tend to count on histopathological findings as the cornerstone in their diagnostic and therapeutic plans.

Previously in this column we reviewed abstracts pointing out the significant percentage of inaccurate diagnoses obtained by needle aspirate cytology and needle-type instrument biopsies of the liver. This review will focus on problems and issues regarding the diagnostic reliability of biopsies of the gastrointestinal tract.

Pathologists' opinions regarding the principal infiltrating cell type

Diagnostic issues

Two recent articles written by veterinary gastroenterologists and pathologists addressed diagnostic issues small animal clinicians should consider when contemplating procedures or acting upon gastrointestinal biopsy findings.

In the first article [Willard, M.D., Lovering, S.L. et al. Quality of tissue specimens obtained endoscopically from the duodenum of dogs and cats. JAVMA, Vol 219, No.4, August 15, 2001] three experienced investigators evaluated the adequacy of 1,335 duodenal tissue samples obtained through endoscopes and submitted to two different laboratories for processing and analysis.

They concluded that 19 to 37 percent of the samples from lab 1 and 23 to 50 percent of the samples from lab 2 were clearly inadequate for diagnostic purposes. The species (canine versus feline) made no difference. The authors noted that the clinicians submitting biopsy samples to lab 1 were senior clinicians at Texas A&M (or individuals trained by these clinicians), and they were very careful to submit their samples with specific orientation on a plastic sponge prior to placing them in formalin.

The clinicians submitting to lab 2 "were from a wide range of training and experience", and samples were submitted floating in formalin rather than with specific orientation. There were some differences in processing technique between labs.

The authors suggested that at least eight individual tissue pieces be submitted when performing endoscopic biopsy of the duodenum on dogs and cats in order to improve the chances of adequate samples.

In the second, related article [Willard, M.D., Jergens, A.E. et al. Interobserver variation among histopathologic evaluations of intestinal tissues from dogs and cats. JAVMA, Vol 220, No. 8, April 15, 2002] the investigators took 14 histologic slides prepared from biopsies of the intestinal tracts of dogs and cats and passed them around for independent evaluation by five veterinary pathologists at four institutions (the investigators specifically chose their 'go to' GI pathologists). Nine slides were made from endoscopically obtained samples and five slides were from full thickness biopsies. Most slides were from animals with clinical signs of GI disease. Some were from "clinically normal" research dogs. The results were eye opening. The quality of all slides were judged adequate or superior by at least four of the five pathologists. However, uniformity of opinion regarding the principal infiltrating cell type was found in only six of the 14 slides (see chart, p. 23S).

Uniformity of opinion regarding the severity of the infiltrate was found in only one of 14 slides. Near uniformity of opinion regarding severity was found in six of 14 slides, and non-uniformity of opinion about severity was found in seven of 14 slides. No pattern or tendency to describe as more or less severe was detected in this group of pathologists. It is notable that two pieces of illeal tissue from the clinically normal dogs were described as having neoplastic infiltrates (lymphoma). The authors concluded that "clinicians must be cautious about correlating clinical signs and histopathologic descriptions of gastrointestinal biopsy specimens."

Consider preliminary diagnostics

In light of this information the practicing clinician should consider the use of preliminary diagnostic tools in animals with GI signs.

Direct fecal smears, flotations, and blood tests such as CBC, biochemical profile, bile acid assays, TLI, Cobalamin, and Folate may define the problem and suggest further diagnostic or therapeutic courses.

Radiographs, ultrasound examination, or an upper GI barium contrast series may clarify the cause of the signs. Empirical 'deworming' and/or a therapeutic trial on a hypoallergenic diet may resolve the signs in some chronic diarrhea cases.

When we do proceed to obtain GI tissue samples, it may be wise to have a veterinarian with formal training in endoscopy perform this particular procedure. Alternatively, surgery and full thickness biopsies should be considered. Attention to your diagnostic lab and pathologist selection is indicated.

Finally, as in all our diagnoses, the history, physical exam, preliminary tests, imaging, and response to therapy must all be considered with the biopsy results in order to deliver our best patient care.

Author's note: It is especially refreshing to see clinicians and pathologists investigating and reporting the issues which relate to the quality and validity of our diagnoses and thus our patient care.