Hayes, Christopher E.; Kawatu, David; Mangray, Shamlal; LeLeiko, Neal S.
Rhode Island Hospital, Providence, RI.
Address correspondence and reprint requests to Neal S. LeLeiko, MD, PhD, MPH 131, Rhode Island Hospital, 593 Eddy St, Providence, RI 02903 (e-mail: Nleleiko@lifespan.org).
Received 14 November, 2011
Accepted 19 January, 2012
The authors report no conflicts of interest.
Objectives: Hirschsprung disease (HD) is defined as the absence of ganglion cells in the Meissner and Auerbach plexuses. Diagnosis depends on demonstrating the absence of ganglion cells in rectal biopsy specimens. Rectal suction biopsy is widely employed as the method of choice in obtaining such specimens. Classically, the diagnosis was made until the 1990s by using the Multipurpose Suction Biopsy Kit, or Rubin Tube. This device was replaced by the Model SBT-100 Suction Biopsy Kit as the exclusive device used to procure rectal tissue. Because the suction devices are known to occasionally yield tissue that is insufficient, the present study was undertaken to determine the efficacy of using this technique to make or exclude the diagnosis of HD.
Methods: The last 50 biopsy sessions using the Multipurpose Suction Biopsy Kit and the first 46 sessions using the Model SBT-100 were included for review.
Results: Both groups had similar yields (24%) of biopsy sessions with insufficient tissue to allow meaningful interpretation. The predictive value of rectal suction biopsy in excluding HD at the first biopsy session was 65%.
Conclusions: We conclude that HD can be excluded with a single rectal suction biopsy 65% of the time. A second biopsy session will exclude the diagnosis in an additional 11% of patients. Both devices yield biopsies of comparable quality and are equally useful in excluding the diagnoses of HD.
Hirschsprung disease (HD) is defined as the absence of ganglion cells in the Meissner and Auerbach plexuses beginning at the anus with variable proximal extension (1). Although many diagnostic methods (eg, acetylcholinesterase staining, barium enema, manometry, immunocytochemical technique) (2–10) have been proposed in the 90 years since the initial description of the disease by Tittel (11), ultimately the diagnosis rests on demonstrating the absence of ganglion cells (1,12). Until the mid-1960s, this necessitated a full-thickness rectal biopsy. In 1965, Dobbins and Bill (13) demonstrated that ganglion cells could be reliably and consistently seen on rectal suction biopsies from normal people that contained only mucosa and submucosa. The clinical usefulness of this finding was that HD could be excluded based on this method of rectal biopsy if ganglion cells were seen. From a practical standpoint, this is sometimes difficult because the suction biopsy specimen often contains only mucosa or small amounts of submucosa and are therefore inadequate to consider the diagnosis when no ganglion cells are seen.
Although rare, known complications of rectal suction biopsy include bleeding and bowel perforation. Rees (14) noted, in a series of 1340 rectal suction biopsies, 3 bowel perforations and 3 rectal hemorrhages. One of the bowel perforations was fatal. Rees implied that keeping the cutting edge of the suction biopsy knife sharp would be important to minimize complications.
For many years, the Multipurpose Suction Biopsy Kit (MSBK) or Rubin Tube (Quinton Instrument Company, Seattle, WA) was the standard instrument used to obtain rectal biopsy specimens. This piece of equipment became unavailable and was replaced >1 decade ago with the Model SBT-100 (Medical Instruments Inc, Hackensack, NJ) as the exclusive device used to procure rectal tissue. There have been no reliable modern reports of the effectiveness of the rectal suction biopsy technique to exclude (or diagnose) HD. This retrospective study was undertaken to compare the success of these 2 devices in obtaining satisfactory biopsy specimens and to examine the usefulness of suction rectal biopsy as a diagnostic tool in the evaluation of suspected HD.
Appropriate permissions to review the pertinent records were granted by the respective institutional review boards. The last 50 biopsy sessions with the MSBK and the first 46 biopsy sessions with the SBT-100 were included for review. The biopsies span a period of 60 months, with the exclusion of a 7-month period during which no record of the biopsy device used was entered. The vast majority of patients were biopsied because of symptoms suggestive of HD. The biopsies were performed by a total of 15 professionals that included attending physicians and fellows in various stages of training. The ages of the patients ranged from 2 days to 18 years with a mean age of 40 months and a standard deviation of 57 months. In accordance with the findings of Aldridge and Campbell (15), biopsies were taken at 3 to 4 cm proximal to the anus using vacuum pressure of 20 inches of mercury. A total of 2 to 3 biopsies were obtained in a circumferential manner so as to obtain appropriate tissue representation. The total number of biopsies obtained from a single patient at any one time constitutes a biopsy session. The tissue was gently removed from the biopsy capsule by a 25-gauge needle, placed on a Teflon gauze, and immediately fixed in a 10% buffered formalin solution. Specimens were embedded in paraffin, and serial sections were done at 5 μm. After discarding the poor-quality ends of the specimen, each biopsy produced 30 to 40 sections of tissue from the middle zone that were stained with Harris hematoxylin and eosin and examined by pediatric pathologists. At least 80 to 100 sections were examined before declaring an absence of ganglion cells.
Based upon the microscopic findings, the biopsies were assigned to 1 of 5 categories: insufficient submucosa for diagnosis; sufficient submucosa, ganglion cells present; sufficient submucosa, ganglion cells absent, hypertrophied trunks; sufficient submucosa, ganglion cells absent; and anorectal mucosa. Category “sufficient submucosa, ganglion cells absent, hypertrophied trunks” is widely thought to be the classical histologic appearance of HD-affected tissue.
In the Multipurpose Tube group, a total of 41 patients were biopsied on 50 separate occasions (biopsy sessions). Twelve biopsy sessions (24%) provided insufficient tissue for diagnosis. Twenty-four patients (58%) had ganglion cells in the first biopsy session. Nine patients underwent a second biopsy session, 5 of them because of persistent symptoms and an insufficient initial biopsy. Of these 5, repeat biopsy revealed ganglion cells in 3. The remaining 2 patients had insufficient tissue on their repeat biopsy as well. An additional 4 patients in the Multipurpose Tube group underwent a second biopsy session. In these 4 patients, the initial biopsy session yielded tissue believed to be of satisfactory quality and size. Although ganglion cells were absent, hypertrophied nerve trunks were not visualized in these samples. These biopsies were considered inconclusive, so these patients underwent a second session in which 3 of the patients were found to have ganglion cells present. The fourth patient was again felt to have sufficient mucosa to allow for diagnosis but ganglion cells were not seen. This patient was subsequently shown to have HD upon full-thickness biopsy. One additional patient with an initial biopsy session that yielded adequate tissue but failed to show ganglion cells underwent laparotomy without a second suction biopsy because of other confirmatory tests. He also was found to have HD.
Two patients had a classic histologic picture of HD in their initial suction biopsy session, both of which were confirmed at subsequent laparotomy without need for a second session. Importantly, of those patients who did not have HD (37 patients), the MSBK correctly identified 65% of them on the first biopsy and an additional 16% at the second biopsy. There were no complications of the procedure in this group.
In the SBT-100 group, a total of 40 patients underwent 46 biopsy sessions (Fig. 1). Eleven biopsy sessions (24%) provided insufficient tissue for diagnosis. Twenty-five patients (62%) had ganglion cells in the first biopsy session. Five patients underwent a second biopsy session, 4 of them because of persistent symptoms and an insufficient initial biopsy. Of these 5 repeat biopsies, ganglion cells were revealed in 2. Two patients again had inadequate biopsies. One of these ended up normal after a third biopsy. The other patient had cessation of symptoms, and no further biopsies were performed. The fifth patient had sufficient tissue on repeat biopsy demonstrating no ganglion cells but with hypertrophied nerve trunks absent. This last patient underwent full-thickness biopsy and ganglion cells were found.
Two patients in the SBT-100 group had a classic histologic picture of HD in their initial suction biopsy session, both of which were confirmed at subsequent laparotomy without need for a second session. In the SBT-100 group, the first biopsy correctly identified 66% of those patients who indeed did not have HD, and an additional 5% at the second biopsy.
Twenty-six biopsy sessions were stratified by age and rated for adequacy of tissue sample. Thirteen of the 16 sessions (81%) in patients younger than 3 years were deemed adequate for pathology review. Seven of the 10 sessions (70%) in patients older than 3 years were deemed adequate for pathology review.
One complication was encountered in this group, in which brisk bleeding occurred in the hours after the biopsy necessitating a packed cell transfusion. Data comparing the 2 devices are summarized in Tables 1 and 2.
Since the landmark article by Dobbins and Bill (13) in 1965 demonstrating the reliability of the rectal suction biopsy technique in the exclusion of HD, this method has become the preferred diagnostic tool for patients suspected as having the disease. The obvious advantage of the suction method is that it spared the patient from a full-thickness biopsy and possible laparotomy with their potential complications; however, clinicians have long known that the technique often provides insufficient tissue for meaningful interpretation, thus leaving the diagnosis in question. Indeed, Dobbins and Bill reported 29 of 309 biopsies (all performed with the multipurpose tube) to have insufficient submucosa (9%). As many as 37% of suction biopsy specimens were found to be inadequate by Campbell, and Noblett (16) found no inadequate specimens when using a device designed specifically for rectal suction biopsies. This device is the predecessor of the SBT-100 suction biopsy gun used in our study. The main difference in the devices was the introduction of a trigger mechanism in the SBT-100 as compared with the Jackson-Negus handle used by Noblett (16).
In our study, we examined not only the sufficiency of the individual biopsies but also the likelihood of a correct diagnosis being made or excluded based upon the information obtained from the biopsy. We then compared the MSBK and the SBT-100 in this regard. As can be seen in Table 1, no obvious differences exist between the 2 instruments in their ability to obtain satisfactory tissue. Insufficient tissue was obtained in 24% of biopsy sessions using the SBT-100 and in 24% of biopsy sessions using the MSBK.
When performing a biopsy on all of the patients suspected as having HD, we found that the diagnosis could be excluded after just 1 biopsy in 65% of those ultimately found not to have the disease. An additional 11% of patients were excluded on subsequent biopsy.
It has been reported that rectal suction biopsies in patients younger than 3 years have a higher success rate than in patients older than 3 years (16–18). Because of the retrospective nature of our study, we had accurate ages on only 26 of our patients. Although our data lacked statistical significance, consistent with the findings of others (16–18), our patients younger than 3 years showed a higher rate of adequate biopsy specimens than did our older patients (81% vs 70%).
Although rectal suction biopsy is the initial method of choice for minimally invasive assessment for HD, the experience of Hirsch et al (19) showed adequate biopsy specimens 93% of the time with the use of jumbo biopsy forceps. No complications in the 167 endoscopic biopsy sessions were noted. This method necessitates sedation in the endoscopy suite, as opposed to the potential bedside nature of rectal suction biopsy.
Clinical suspicion on the part of the clinician coupled with a satisfactory specimen showing only the absence of ganglion cells warranted a full-thickness biopsy. Of 7 patients referred for full-thickness biopsy, 1 was found not to have HD. Four of these 7 patients had the “classic” histologic picture of HD; the rest had only absent ganglion cells.
In our study, of 6 patients proven to have HD at laparotomy, 4 had absence of ganglion cells and thickened nerve trunks and the other 2 had only absent cells. Of the patients selected for biopsy, 7% proved to have the disease. This is an acceptable yield for a procedure that carries minimal risk. Little data exist with which to compare these results.
One complication occurred in which a patient biopsied with the SBT-100 experienced brisk bleeding requiring transfusion. Review of the sections showed no large vessels, and no coagulopathy was present. This represents a complication rate of 1%.
In conclusion, we found that rectal suction biopsy excluded the diagnosis in 76% of patients without HD. In patients with the disease, suction biopsy either diagnosed or provided information leading to diagnosis in all of the patients. In addition, no differences exist between the MSBK and the SBT-100 with regard to their ability to obtain satisfactory specimens.
1. Whitehouse FR, Kernohan JW. Myenteric plexus in congenital megacolon: study of 11 cases. Arch Intern Med 1948; 82:75–111.
2. Kekomaki M, Rapola J, Louhimo I. Diagnosis of Hirschsprung's disease. Acta Paediatr Scand 1979; 68:893–897.
3. Lawson JON, Nixon HH. Anal canal pressure in the diagnosis of Hirschsprung's disease. J Pediatr Surg 1967; 2:544–545.
4. Aaronson I, Nixon HH. A clinical evaluation of anorectal pressure studies in the diagnosis of Hirschsprung's disease. Gut 1972; 13:138–141.
5. Meier-Ruge W, Lutterbeck PM, Herzog B, et al. Acetylcholinesterase activity in suction biopsies of the rectum in the diagnosis of Hirschsprung's disease. J Pediatr Surg 1972; 7:11–21.
6. Morikawa Y, Donahoe P, Hendren WH. Manometry and histochemistry in the diagnosis of Hirschsprung's disease. Pediatrics 1979; 63:865–871.
7. Low PS, Quak SH, Prabhakaran K, et al. Accuracy of anorectal manometry in the diagnosis of Hirschsprung's disease. J Pediatr Gastroenterol Nutr 1989; 9:342–346.
8. Schofield DE, Devine W, Yunis EJ. Acetylcholinesterase-stained suction rectal biopsies in the diagnosis of Hirschsprung's disease. J Pediatr Gatroenterol Nutr 1990; 11:221–228.
9. Hirose R, Hirata Y, Yamada T, et al. The simple technique of rectal mucosal biopsy for the diagnosis of Hirschsprung's disease. J Pediatr Surg 1993; 28:942–944.
10. Ikawa H, Kim SH, Hendren H, et al. Acetylcholinesterase and manometry in the diagnosis of the constipated child. Arch Surg 1986; 121:435–438.
11. Tittel K. Ueber eine angeborene missbildung des S dickdarmes. Wien Klin Wschr 1901; 14:903.
12. Bodian M, Stephens FD, Ward BCH. Hirschsprung's disease. Lancet 1950; 1:19–22.
13. Dobbins WO, Bill AH. Diagnosis of Hirschsprung's disease excluded by rectal suction biopsy. N Engl J Med 1965; 272:990–993.
14. Rees BI, Azmy A, Nigam M, et al. Complications of Rectal suction biopsy. J Pediatr Surg 1983; 18:273–275.
15. Aldridge RT, Campbell PE. Ganglion cell distribution in the normal rectum and anal canal: a basis for the diagnosis of Hirschsprung's disease by anorectal biopsy. J Pediatr Surg 1968; 3:475–490.
16. Campbell PE, Noblett HR. Experience with rectal suction biopsy in the diagnosis of Hirschsprung's disease. J Pediatr Surg 1969; 4:406–414.
17. Croffie JM, Davis MM, Faught PR, et al. At what age is a suction rectal biopsy less likely to provide adequate tissue for identification of ganglion cells? J Pediatr Gastroenterol Nutr 2007; 44:198–202.
18. Alizai NK, Batcup G, Dixon MF, et al. Rectal biopsy for Hirschsprung's disease: what is the optimum method? Pediatr Surg Int 1998; 13:121–124.
19. Hirsch BZ, Angelides AG, Goode SP, et al. Rectal biopsies obtained with jumbo biopsy forceps in the evaluation of Hirschsprung disease. J Pediatr Gastroenterol Nutr 2011; 52:429–432.