Hirschsprung Disease and Use of Calretinin in Inadequate Rectal Suction Biopsies

After institutional review board approval, an electronic search of the pathology records was performed to identify biopsies “inadequate for diagnosis” between January 2000 and January 2012 at the Cleveland Clinic, Cleveland, Ohio. These biopsies were considered inadequate because they contained no ganglion cells in H&E-stained sections and either anal mucosa or limited to no submucosa was present (Figure, A). A total 20 cases were obtained, and the original H&E-stained slides were reviewed to ensure that columnar-lined mucosa (lamina propria and muscularis mucosae) was present. Owing to lack of clinical follow-up (1 patient moved to another state), or the presence of anal squamous mucosa only (2 cases), the final number of selected cases was reduced to 17. Subsequent pathology reports and the patients' electronic medical records were reviewed to determine clinical outcome. Ten resection specimens of HD were retrieved from the Cleveland Clinic files. Distal (aganglionic) plus proximal (ganglionated) bowel was evaluated with calretinin IHC in these specimens. This allowed for confirmation that calretinin staining properly functioned in our laboratory.

Next, paraffin-embedded, 4-μm-thick sections and available unstained slides from the 17 selected cases were used for calretinin IHC (Figure, B). Immunohistochemical staining was performed with the Benchmark XT, an automated immunostainer from Ventana Medical System (Tucson, Arizona). The specific protein-antibody complexes were located by using a biotin/streptavidin-horseradish peroxidase/diaminobenzidine tetrahydrochloride (DAB) detection kit (i-View DAB, catalog No, 760-050, Ventana) to reduce nonspecific staining caused by endogenous biotin in tissues. Immunohistochemistry was performed by using a rabbit polyclonal antibody from Invitrogen (catalog No. 08–1211, Grand Island, New York) at a dilution of 1:40. Epitope retrieval process lasted for 30 minutes at 95°C in high pH buffer (CC1 mild), and incubation was carried out for 60 minutes at 37°C. Calretinin slides were evaluated independently by the 2 authors and without knowledge of clinical outcome. Calretinin immunostaining was scored as either positive or negative. The location of the staining (lamina propria, muscularis mucosae, and submucosa) was also registered (Table 1). Fisher exact test was used in our study to compare the calretinin stain results with the presence or absence of HD.

The 17 patients ranged in age from 3 days to 16 years. There were 7 females and 10 males. The number of H&E levels per case ranged from 1 to 75 (mean, 14.1; standard deviation, 22.2). Three categories of follow-up emerged: (1) definitely not HD, confirmed by subsequent RSB with unequivocal ganglion cells (7 cases); (2) definitely HD, confirmed by surgical resection specimen (5 cases); and (3) clinically not HD, substantiated by numerous notes within the medical record without mention of persistent constipation or HD (5 cases with a follow-up ranging between 2.8 to 11.1 years; mean, 7.1 years).

For all patients, absence or presence of calretinin was correlated with the pathologic or clinical follow-up. Twelve cases were positive for calretinin. In 4 cases, the calretinin stain was present in lamina propria, muscularis mucosae, and submucosa. In 5 cases, the stain was present in lamina propria and muscularis mucosae. In 3 cases, calretinin stain was observed in lamina propria alone. A total of 5 cases were negative for calretinin. Cases 1, 5, 7, 10, and 11 lacked muscularis mucosae on deeper levels for which calretinin IHC was performed (Table 1).

The concordance between calretinin staining and presence or absence of HD in clinical follow-up was statistically significant (P = .004, exact Fisher test) (Table 2).

According to a recent consensus article by the Gastro 2009 International Working Group,¹¹  diagnostic rectal biopsies should be taken 1 to 2.5 cm proximal to anorectal squamocolumnar junction, should measure at least 3 mm, and should include at least one-third of the submucosa.

Having adequacy standards implies that these standards will not always be met. Rectal suction biopsies are not an exception, and up to 17%¹⁰  of biopsies may be inadequate, particularly in older children for whom the submucosa tends to be more fibrous and from whom less tissue is obtained during the suctioning procedure. In our study, 13 of 17 patients were older than 6 months. For this specific population, some groups recommend near-full-thickness biopsies, given the increased risk of inadequate RSB.

Analysis of acetylcholinesterase (AChE) activity is one of the most widely used ancillary techniques in the workup of HD. The number and the thickness of AChE-positive nerve fibers are increased in the muscularis mucosae and lamina propria in rectal biopsies of most patients with HD.¹²  Although used for decades,¹³  AChE has several disadvantages. Acetylcholinesterase histochemistry requires the use of frozen tissue, interrupting the usual workflow of a surgical pathology laboratory. In addition, AChE activity fades after time, requiring continual replenishment of HD control tissue. Moreover, interpreting AChE patterns requires experience, and AChE staining has been associated with relatively high rates of interobserver disagreement^7,8  as well as false-positive results.

Calretinin stain has been shown to be superior to AChE stain in 2 previous studies,^7,8  with less interobserver disagreement and no false-positive interpretations. Calretinin holds many advantages over AChE. It can be performed on paraffin-embedded tissue, whereas frozen tissue is necessary for AChE. Most IHC laboratories have calretinin available. Furthermore, studies have shown that calretinin is interpreted more reliably and seems to require less observer experience than AChE.^7,8 

The current study showed that the use of calretinin in inadequate biopsies helps identify the presence or absence of HD. The diagnosis of HD is still centered on evaluating the absence of ganglion cells. In inadequate biopsies, however, this is not possible. In this scenario, calretinin IHC results may be very helpful in adequately triaging patients. We see this as particularly helpful in reducing additional RSBs or even more invasive procedures such as full-thickness biopsies.¹⁴ 

We offer 2 points of caution in interpreting our results. First, loss of IHC is best interpreted with positive staining of internal control, which is often satisfied by identifying calretinin-positive mast cells. Occasionally, no positive control is identified and in these cases, some caution is warranted in interpreting the calretinin staining. Second, although 5 of our cases were interpreted correctly without the presence of muscularis mucosae, a minority of cases have show only calretinin-positive nerve fibers in the muscularis mucosae.⁸  Therefore, false-negative calretinin results are possible in biopsies lacking muscularis mucosae.

The emerging literature of calretinin in the setting of HD is somewhat limited, but quite encouraging. These studies have shown calretinin utility not only in HD specimens⁶  but also in adequate rectal suction biopsies.^4,7,8  Furthermore, Kapur et al⁸  reported their experience with the use of calretinin stain in a group of 7 inadequate biopsies (3 of the cases due to insufficient submucosa and 4 with biopsies including anorectal zone) from patients with proved absence of HD. They found perfect correlation with calretinin IHC and HD status as long as columnar mucosa with muscularis mucosae was sampled. Our series of 17 inadequate biopsies strengthens these findings.

In the future, calretinin IHC may help pathologists and histology laboratories to save time and reduce the seemingly innumerable histologic levels performed in the workup of HD; however, this study was not designed to address the potential workload reductions by adding calretinin. To examine the impact of calretinin on reducing the number of histologic levels, a study needs to begin with adequate aganglionic RSBs that integrate calretinin IHC before exhausting the block. Since we have integrated calretinin in our protocol only during the last 2 years, we are still acquiring data on this subject.

In conclusion, calretinin provides a very reliable adjunctive test in the evaluation of RSBs for HD and, in our opinion, should be routinely used as an additional tool, similar to presence of hypertrophic nerves or AChE histochemistry, particularly because surgeons are often looking for an additional corroborative histologic finding to aganglionosis before undertaking surgery.

The authors would like to thank Amy Posch, BS, MT, for her technical expertise in perfecting the calretinin stain. This study would not have been finalized without her commitment and her determination to improve the calretinin immunostain in our laboratory.