Find best premium and Free Joomla templates at

Correlation of Radiographic and Histologic Transition Zones in Hirschsprung's Disease

Ike Njere¹³, Bijan Hedayati², Reza Salari², Dorothy Kufeji¹³, Alireza S. Keshtgar¹³
¹Department of Pediatric Surgery, University Hospital Lewisham, London, UK
²Department of Radiology, University Hospital Lewisham, London, UK
³Department of Pediatric Surgery, Evelina Children Hospital, Guy’s and St Thomas’, London, UK



Purpose: Hirschsprung’s disease (HD) is a common paediatric surgical condition with an incidence of 1 in 5000 live births. There has been renewed interest in the role of contrast enema for predicting the level of aganglionosis with single stage endoanal pullthrough in HD. Few studies have evaluated the accuracy of contrast enema in identifying the transition zone. The aim of this study was to ascertain the correlation between the radiographic and histologic transition zones (HTZ) in Hirschsprung’s disease and to determine the sensitivity and specificity of contrast enema in identifying the transition zone.

Material and method: We retrospectively studied 61 patients with Hirschsprung’s disease, presenting between January 2006 and October 2010. Case notes, radiology and histology reports were reviewed. Correlation was defined as the Radiographic Transition Zone (RTZ) in the same or adjoining segment of bowel to the transition zone established by histology. All patients were initially managed by rectal washouts prior to obtaining contrast enema.

Results: Of the 61 patients, there was complete data in 52/61. Of these 22 had contrast enema. Seventeen of them were males. Majority were term babies with a median age at diagnosis of 10 days. An RTZ was clearly defined in 18 of 22 that had a contrast enema. In 16/18 patients there was concordance between the RTZ and HTZ. Contrast enema had a sensitivity of 93% and specificity of 50% for identifying the RTZ in the recto-sigmoid region. The discordant rate was 11.1% (2/18).

Conclusions: In the majority of patients, the transition zone can be predicted by contrast enema (88% concordance). The radiological appearance is not affected by prior rectal washout. Long segment disease and catheter tip beyond the anal canal contribute to discordance.

Keywords: Hirschsprung’s, contrast enema, histology, concordance



Alireza S. Keshtgar
Department of Pediatric Surgery
Evelina Children Hospital, Guy’s and St Thomas’ NHS foundation Trust,
Westminster Bridge Road,
London SE1 7EH, United Kingdom
Tel: 020 7188 4674
Fax: 020 7188 4612
Email: This email address is being protected from spambots. You need JavaScript enabled to view it.



Hirschsprung’s disease is a common paediatric surgical condition with an incidence of 1 in 5000 [1]. The pathology is characterised by absence of ganglia and presence of hypertrophic nerve bundles in the bowel wall. The deficiency begins at the internal anal sphincter and extends proximally to a variable extent [1]. Associated are abnormalities of the adrenergic and cholinergic nervous system.

The result is a tonic contraction of the bowel which is therefore unable to propagate peristaltic waves. The spasm may be due to excess acetylcholine release, with a resultant abundance of acetylcholinesterase activity. This is noted in the hypertrophic nerve bundles. The smooth muscle cells also have increased sensitivity to acetylcholine [1]. The normal ganglionic bowel proximal to the aganglionic segment becomes dilated and the funnel shaped intervening hypoganglionic bowel segment between the proximal dilated segment and distal contracted segment is the transition zone.

Rectal biopsy for diagnosis of Hirschsprung’s disease was first recommended by Swenson et al. in 1955 [2]. Rectal biopsy is the current gold standard for diagnosing Hirschsprung’s disease [3]. Levelling biopsies are obtained at the time of stoma creation or at the definitive pull-through procedure to determine the level of aganglionosis and the transition zone. A contrast enema is a useful tool in the management of Hirschsprung’s disease. Swenson et al. in 1949 described the classic transition zone as occurring between proximal dilated bowel and narrow aganglionic distal bowel [4]. It fell into disuse but there has been resurgence in its use because of the advent of single stage endoanal pull through procedures. These procedures make it a useful tool in identifying the Radiographic Transition Zone (RTZ) which aids in choice of procedure and planning the operation [3].

There is therefore the need to have an accurate interpretation of the RTZ zone to avoid difficulties during the procedure. It is desirable to have correlation of the RTZ and the HTZ. The concordance rate between the RTZ and the HTZ has been reported to be 62.5-88% in all segments and 75-90% in the rectosigmoid segments [5, 6]. A lower concordance rate of 25-31% was reported for longer segment Hirschsprung’s disease [5, 6].

The aim of the study was to look at the correlation of the radiographic transition zone (RTZ) with the histologic transition zone (HTZ) and to determine the sensitivity and specificity of contrast enema in identifying the transition zone.

Material and Method

This was a retrospective study of 61 patients with Hirschsprung’s disease, who were treated in our centre (a tertiary referral centre), between January 2006 and October 2010. We obtained approval from the research and audit department of University Hospital Lewisham, NHS Trust, London.

We had complete data on 52 patients of which 22 had a contrast enema, which is reported in this study. Case notes, radiology and histology reports were reviewed. All the contrast enemas were reviewed by one experienced consultant radiologist.

Correlation was accepted if the RTZ was in the same segment of bowel or in an adjoining segment to the transition zone established by histology.

Data was recorded on an Excel spreadsheet. Statistical analysis was carried out with Stata software version 8.2 (Statacorp Lakeway drive, College Station, Texas, USA.).


During the study period, 61 patients with Hirschsprung’s disease were treated at our centre. Fifty-two of them had complete data. Of the 52, 22 had a contrast enema before the definitive pull-through procedure. There were 17 males and 5 females. The median gestational age and birth weight were 39 weeks (range 35 – 42) and 3.4 kilograms (kg) (range 2.15 – 4.9) respectively. The median age at diagnosis was 10 days (range 3 – 2190) while the median weight and age at surgery were 4.9 kg (range 3.4-18.4) and 2 months (range 0.36 – 72) respectively.

The histologic transition zone was in the recto-sigmoid in 19 patients (86.4%), in the descending colon in 1 patient (4.5%) and in the distal ileum (total colonic) in 2 patients (9.1%).

Of the 22 patients that had a contrast enema, 18 patients had an RTZ defined while 4 patients did not. The histologic transition zones in these 4 patients were in the mid sigmoid, mid/distal sigmoid, recto-sigmoid and proximal/mid rectum. There was an 88% (16/18) concordance between the RTZ and the HTZ. The sensitivity and specificity of contrast enema for identifying the transition zone in the recto-sigmoid region was 93% and 50% respectively.

Most patients had a Duhamel pull-through and 68.2% (15/22) had a primary pull-through procedure (Table 1).

Table 1. Number of patients with staged versus primary procedure





















Of the 19 patients that had a Duhamel pullthrough,18 had recto-sigmoid disease. The only patient that had a Swenson’s pull-through had a total colonic Hirschsprung’s disease, while one of the 2 patients that had a Soave pull-through had total colonic disease.


Hirschsprung’s disease constitutes 1.2 to 10% of all paediatric surgical admissions and therefore an important cause of paediatric surgical bed occupancy [7, 8]. The diagnosis of Hirschsprung’s disease is made from a rectal biopsy. Zuelzer and Wilson first established the absence of ganglion cells as the pathologic basis of this disease [9]. They noted absence of ganglion cells in a portion of the bowel which appeared to correlate with the level of obstruction in 6 of their cases. Swenson et al. then recommended rectal biopsy for the diagnosis of Hirschsprung’s disease [2]. Ehrenpreis was the first to report on the roentgenographic findings in Hirschsprung’s disease in newborns [10]. Serial barium enema studies were carried out in 4 of his cases and the “typical roentgen picture” was said to develop between the ages of 18 days and 3 1/2 months.

Swenson et al. had described the classic transition zone as occurring between proximal dilated bowel and narrow aganglionic distal bowel [4]. McDonald and Evans stressed that the diagnosis in infancy depended on the contrast enema but that this may be negative in early “mild” cases [11]. They felt that abnormal stasis of the barium in the colon after the barium enema was the earliest roentgenographic sign of the disease. However, de Lorjn et al. in a systematic review noted a sensitivity of 70% and a specificity of 83% with contrast enema in diagnosing Hirschsprung’s disease [12]. The contrast enema is therefore most useful for indicating the proximal extent of the disease by identifying the transition zone. Its correlation with the transition zone identified by histology is a validation of the ability of the contrast enema to guide us in planning a surgical procedure. With the advent of single stage endoanal pull-through in HD, the importance of this correlation has been recognised. Very few studies have looked at this correlation with the concordance reported as ranging from 62.5-88% in all segments [5, 6].

In this retrospective study, we looked at the correlation between the RTZ and the HTZ on our patients with HD and also determined the sensitivity and specificity of contrast enema in identifying the transition zone in our centre.

Procedures that are used in treating this condition include Swenson’s pull-through (1948), Duhamels’s pull-through (1956) and Soave’s endorectal pull-through (1964) amongst others [13]. More recent additions to the surgical armamentarium for this condition include laparoscopic assisted pull-through procedures and transanal endorectal Soave pull-through [13]. The desire for minimally invasive procedures has made the transanal endorectal pullthrough an attractive option. De la Torre and Ortega first described the transanal pull-through procedure in 1998 [14]. Many surgeons have since adopted the procedure to treat Hirschsprung’s disease. Whether the transanal procedure should be carried out without first determining the level of aganlionosis is a contentious point. Some authors believe that the transanal Soave is not ideal for long segment colonic disease and that this procedure however is more suited for rectosigmoid disease [15]. Langer J.C et al. in 1999 had suggested that a completely transanal approach was not appropriate in a child in whom the level of aganglionosis extended proximal to the sigmoid colon because of difficulties in mobilizing the colon [16]. Podevin et al. noted that evaluating the level of aganglionosis was a difficulty encountered by surgeons carrying out this procedure [17].

The level of aganglionosis could be determined preoperatively by a contrast enema or intraoperatively by levelling biopsies. Intraoperative levelling biopsies could be performed by laparotomy or by laparoscopy. However the transanal Soave procedure aims to avoid abdominal incisions for better cosmetic and quicker recovery. Preoperative identification of the transition zone by contrast enema would therefore be more desirable. Berman in 1956 reported that contrast enema was able to identify the transition zone in 80% (15 /19) of her patients with Hirschsprung’s disease [18]. Proctor et al. reported that preoperative contrast enema identified the transition zone in 89% (67/75) and failed to identify the transition zone in 11% of their patients [6]. Our study showed a similar finding with the transition zone identified in 81% (18/22) of our patients and not identified in 4 patients. The reason the transition zone was not identified in these patients could be attributed to technical problems with performing the radiologic study. The histologic transition zones in these 4 patients were in the mid sigmoid, mid/distal sigmoid, rectosigmoid and proximal/mid rectum. Berman had warned against the temptation of inserting the catheter used in performing the study too far into the rectum [18]. She noted that a markedly contracted rectum could make inflow of barium difficult and attempting to push the catheter further would miss the transition zone. She also identified long segment disease as another reason for not visualizing the transition zone. Sane and Girdany reported a normal contrast study in 22% (17/76) infants and a 15% (11/76) false impression of transition zone in infants with total aganglionosis coli (TAC) [19].

The accuracy of the radiographic transition zone in predicting the histologic transition zone has not been well documented. Rosenfield et al. in 1984 reported a diagnostic sensitivity and specificity of 62% and 100% respectively when an RTZ was present in the rectosigmoid and a sensitivity of 70% and a specificity of 80% with all segments [20]. They had a 100% concordance rate as the RTZ correlated with the HTZ in all 27 patients compared. However other investigators have not obtained the same result. Filling of the proximal aganglionis bowel segment with retained stool has been reported as a reason for a more distal appearing RTZ [21]. The 11 of 76 patients with TAC reported by Sane and Girdany to have a false impression of transition had it in an area of the colon well below the HTZ [19].

In our study, 16 of the 18 patients that had an RTZ identified on contrast enema, had correlation with the HTZ, therefore giving a concordance rate of 88%. We noted a concordance rate of 93% (14/15) in the rectosigmoid segment. This is similar to that reported by Proctor et al in 2003 [6]. They reported that in 89% (67/75) of their patients the contrast enema showed an RTZ suggestive of Hirschsprung’s disease. The pathologic and radiographic transition zones were concordant in 88% (59/67) of their patients. They noted a higher rate of concordance of 90% if the RTZ was in the rectosigmoid region. Jamieson et al. in 2004 reported a lower overall concordance rate of 62.5% and a concordance rate of 75% in the short segment disease [5]. They divided their study group into short segment HD and long segment HD using the distal descending colon as cut off. This was because the emphasis was on short segment disease as the transanal Soave procedure is most suited to this.

Different investigators have obtained different values for contrast enema diagnosis of Hirschsprung’s disease in the rectosigmoid region. Our study noted a sensitivity of 93% and specificity of 50% for contrast enema diagnosis of Hirschprung’s disease in the rectosigmoid region. It also indicated a positive predictive value of 93% and negative predictive value of 50%. This is similar to the findings of Jamieson et al. who obtained a sensitivity of 92% and a specificity of 75% for short segment HD [5]. They also obtained a positive predictive value of 92% and a negative predictive value of 75%. Rosenfield et al. however reported a much lower sensitivity of 54% and 100% specificity when a rectosigmoid RTZ was present [20]. They obtained a positive predictive value in the rectosigmoid of 100% and a negative predictive value of 54%. The reason for their different values is unclear.

We did not find that rectal manipulation in the form of digital rectal examinations or rectal washout had any bearing on the findings in this study. The same observation was made by Rosenfield et al. [20].

In summary our study suggests that 1 out of 3 patients with long segment HD would have the transition zone not correctly identified by contrast enema. However the high sensitivity for the rectosigmoid disease still makes it a useful procedure as this is the segment most amenable to the transanal Soave procedure. If longer segment disease is suspected laparoscopy and laparoscopic mobilization of bowel would be a useful option. It has to be emphasised that care must be taken in performing the contrast enema to obtain optimal results.




1. Whan E.B: Hirschsprung disease, in Burge D.M, Griffiths D.M, Steinbrecher H.A, Wheeler R.A (eds): Paediatric Surgery. London, Hodder Arnold, 2005, pp147-153.

2. Swenson O, Fisher JH, MacMahan HE: Rectal biopsy as an aid in the diagnosis of Hirschsprung’s disease. N Engl J Med 1955; 253: 632-635.

3. Martucciello G, Pini Prato A, Puri P et al: Controversies concerning diagnostic guidelines for anomalies of the enteric nervous system: A report from the fourth International Symposium on Hirschsprung’s disease and related neurocristopathies. J Pediatr Surg 2005; 40: 1527-1531.

4. Swenson O, Neuhauser E.B.D, Pickett L.K: New concepts of the etiology, diagnosis and treatment of congenital megacolon (Hirschsprung’s disease). Pediatrics 1949; 4: 201-209.

5. Jamieson D.H, Dundas S.E, Belushi S.A et al: Does the transition zone reliably delineate aganglionic bowel in Hirschsprung’s disease? Pediatr Radiol 2004; 34: 811-815.

6. Proctor M.L, Traubici J, Langer J.C et al: Correlation between Radiographic Transition Zone and Level of Aganglionosis in Hirschsprung’s Disease: Implications for Surgical Approach. J Pediatr Surg 2003; 38: 775-778.

7. Bickler S.W, Sanno-Duanda B: Epidemiology of paediatric surgical admissions to a government referral hospital in the Gambia. Bulletin of the World Health Organization 2000; 78 (11): 1330-1336.

8. Nandi B, Mungongo C, Lakhoo K: A comparison of neonatal surgical admissions between surgical departments in Africa and Europe. Pediatr. Surg. Int 2008; 24: 939-942.

9. Zuelzer W.W, Wilson J.L: Functional intestinal obstruction on a congenital neurogenic basis in infancy. Am J. Dis. Child 1948; 75:40.

10. Ehrenpreis T: Megacolon in the newborn; a clinical and roengenological study with a special regard to the pathogenesis. Acta chir. Scandinav 1946; 94: Suppl. 112, p. 1

11. McDonald R.G, Evans W.A Jr: Hirschsprung’s disease. Roentgen diagnosis in infants. Am. J. Dis. Child 1954; 87: 575.

12. De Lorijn F, Kremer L.C.M, Reitsma J.B et al: Diagnostic Tests in Hirschsprung Disease: A Systematic Reviw. J Pediatr Gastroenterol Nutr 2006; 42 (5): 496-505.

13. Burki T, Sinha C.K, Yamataka A: Hirschsprung’s Disease, in Sinha C.K, Davenport M (eds): Handbook of Pediatric Surgery. London, Springer-Verlag, 2010, pp117-124.

14. De la Torre-Mondragon L, Ortega-Salgado JA : Transanal endorectal pull-through for Hirschprung’s disease. J Pediatr Surg 1998; 33:1283-1286.

15. Hoehner J.C, Ein S.H, Shandling B et al: Long term mobidity in total colonic aganglionosis. J. Pediatr Surg 1998; 33:961-965.

16. Langer J.C, Minkes R.K, Mazziotti M.V et al : Transanal One-Stage Soave Procedure for Infants With Hirschprung’s Disease. J. Pediatr Surg 1999; 34(1):148-152.

17. Podevin G, Lardy H, Azzis O, Branchereau S, Petit T, Sfeir R, Well D, Heloury Y, Fremond B: Technical problems and Complications of a Transanal Pull-Through for Hirschsprung’s Disease. Eur J Pediatr Surg 2006; 16: 104-108.

18. Berman C.Z: Roentgenographic manifestations of congenital megacolon (hirschsprung’s disease) in early infancy. Pediatrics 1956; 18(2): 227-238.

19. Sane S.M, Girdany B.R: Total aganglionosis coli: Clinical and roentgenographic manifestations. Radiology 1973; 107: 397-404.

20. Rosenfield N.S, Ablow R.C, Markowitz R.I et al: Hirschsprung disease: Accuracy of the barium enema examination. Radiology 1984; 150: 393-400.

21. Evans W.A, Willis R: Hirschsprung’s disease: The roentgen diagnosis in infants. AJR 1957; 78: 1024-1048.