Surgery for Pediatric Anorectal Malformation (Imperforate Anus)

Updated: Jun 17, 2022
  • Author: Richard J Wood, MD, MBChB, MRCS(Eng), Dip PEC(SA), FCPaedSurg(SA); Chief Editor: Eugene S Kim, MD, FACS, FAAP  more...
  • Print

Practice Essentials

Anorectal malformations (ARMs) comprise a wide spectrum of anomalies that affect boys and girls and can involve the distal anus and rectum, as well as the urinary and genital tracts. Malformations range from minor and easily treated defects that carry an excellent functional prognosis to complex defects that, despite successful treatment, are often associated with other anomalies and carry a poor functional prognosis.

Throughout the centuries, doctors have seen and attempted to treat infants born with imperforate anus. Given that few patients were described in the early literature, most are assumed to have died without treatment. Paulus Aegineta, in the 7th century CE, recorded the earliest account of successful surgery for imperforate anus. He suggested rupturing an obstructing membrane with a finger or knifepoint and then dilating the tract until healing was complete. This approach was used for many years.

Almost 1000 years later, in 1660, Scultet used dilatation to treat an infant with anal stenosis. In 1676, Cooke used incision and dilatation and advised care of the sphincter muscles. In 1787, Bell suggested using a midline perineal incision to find the bowel. In 1783, acting on Littre's suggestion from 1710, Dubois performed an inguinal colostomy for imperforate anus. Other surgeons followed suit, but almost all of the infants died; thus, colostomy remained unpopular and a procedure only of last resort.

In 1835, Amussat described formal perineal proctoplasty (ie, mobilization of the bowel through a perineal incision and suturing to the skin). This technique gained rapid acceptance. Strictures were less common than observed in earlier procedures. In addition to Amussat, Dieffenbach described anal transposition (1826); Chassaignac used a probe through a stoma to guide the perineal dissection (1856); and Leisrink (1872), McLeod (1880), and Hadra (1884) recommended opening the peritoneum if the bowel was not encountered from below.

In 1930, Wangensteen and Rice first advocated imaging to delineate the abnormality. Single-stage abdominoperineal procedures became widely used after reports by Rhoads, Pipes, Randall, Norris, Brophy, and Brayton (1948-1949). In 1953, Stephens described this procedure with specific emphasis on preservation of the puborectalis muscle. This surgical approach and its modifications were the standard until 1980.

In 1980, the surgical approach to repairing ARMs altered dramatically with the introduction of the posterior sagittal approach by Peña and Devries. This approach allowed pediatric surgeons to obtain a clear view of the anatomy of ARMs and to repair them under direct vision, with better visualization and understanding of the anatomic relations than previous approaches allowed. Georgeson subsequently added laparoscopy so that any abdominal dissection could avoid an incision.

Since the introduction of these two approaches, there has been considerable debate regarding their efficacy. There has been a growing consensus that some malformations are best approached transabdominally (via laparoscopy or open surgery, depending on available expertise and equipment) and others posterior-sagittally. 



Several classifications have been used for ARMs in the past. At present, it is widely accepted that the Krickenbeck classification from 2005, which was a modification of the previous Peña classification, should be used to describe ARMs (see Table 1 below). This is a departure from the previous Wingspread classification (1984) insomuch as it relies on the specific anatomy of each defect rather than on broad and unhelpful terms such as high, intermediate, and low, which have only a limited bearing on prognosis.

Table 1. Krickenbeck Classification of Anorectal Malformations (Open Table in a new window)

Major Clinical Groups Rare/Regional Variants
Perineal (cutaneous) fistula Pouch colon

Rectourethral fistula

  • Prostatic
  • Bulbar
Rectal atresia/stenosis
Rectovesical fistula Rectovaginal fistula
Vestibular fistula H fistula
Cloaca Others
No fistula  
Anal stenosis  

The goal of this type of classification system is to define the anatomy of malformations reliably. This is especially important for comparing data from different sites, both nationally and internationally. Making meaningful comparisons is greatly facilitated by using clearly defined anatomic definitions. A 2019 publication by Halleran et al sought to further define the anatomy of rectourethral malformations. [1]


Pathophysiology and Etiology

A sound understanding of the anatomy is helpful to prevent damage to important structures during the surgical repair and to preserve the best potential for fecal and urinary continence.

Anatomic visualization has allowed surgeons to eliminate many previous misconceptions. For instance, the previous classification of these defects into high, intermediate, and low malformations was a misleading oversimplification that did not adequately describe the spectrum of anorectal anomalies.

Improved imaging techniques and a more thorough knowledge of the anatomy of the pelvic structures at birth have refined diagnosis and early treatment. Analysis of large series of patients has allowed better prediction of associated anomalies and functional prognosis.

The surgeon’s primary concerns in correcting these anomalies are to accomplish a successful anatomic reconstruction and to achieve bowel and urinary control. Problems with sexual function and fertility must also be considered.

Early diagnosis, treatment of associated anomalies, and efficient and meticulous surgical repair provide patients the best chance for a good functional outcome.

Some patients experience fecal and occasional urinary incontinence despite excellent anatomic repair. Associated problems (eg, poorly developed sacrum, pelvic musculature, and pelvic nerve roots, as well as tethered cord or myelomeningocele) likely contribute to an inability to achieve continence. For patients with true fecal incontinence, an effective bowel management can provide social cleanliness and effective evacuation of stool. This in turn may improve urinary function, reduce urinary tract infections (UTIs), and improve quality of life. (See Bowel Management.)

Although the etiology remains unknown, a slight genetic predisposition appears to exist.



ARMs occur in approximately one out of every 3000-5000 births and are slightly more common in males, [2, 3]  with a 1% risk that a family will have a second child with an ARM. [4, 5]

A rectourethral fistula is most common in males, and a rectovestibular fistula is most common in females. Having no fistula at all is rare (5% of patients) and is associated with Down syndrome. [6]

In the past, cloaca was considered a rare defect, whereas rectovaginal fistula was commonly reported. In fact, the converse is true: Cloacas are the third most common defect in females, after vestibular and perineal fistulas. A true rectovaginal fistula is rare (< 1% of cases). [7]  Incorrect diagnosis in such a case leads to surgical treatment in which only the rectal component is repaired and the patient is left with a a persistent urogenital sinus. [8]



In evaluating the results of treatment of ARMs, patients should not be grouped into the traditional high, intermediate, and low categories from the 1984 Wingspread classification. This classification is flawed in several ways. For instance, within the high group, different ARMs have different treatments and carry different prognoses (eg, rectoprostatic fistula compared with rectobladderneck fistula). Both of these would be considered high in the Wingspread classification; however, the two malformations are so different that they should not be grouped together.

As stated above, an anatomically based grouping is of more clinical value, and this is provided by the Krickenbeck classification (see Table 1 below). [9]

Table 1. Krickenbeck Classification of Anorectal Malformations (Open Table in a new window)

Major Clinical Groups Rare/Regional Variants
Perineal (cutaneous) fistula Pouch colon

Rectourethral fistula

  • Prostatic
  • Bulbar
Rectal atresia/stenosis
Rectovesical fistula Rectovaginal fistula
Vestibular fistula H fistula
Cloaca Others
No fistula  
Anal stenosis  

The functional results of repair of anorectal anomalies have improved significantly since the advent of the posterior sagittal approach. However, the results of this approach are difficult to compare with those of other methods because terminology and classification are inconsistent.

Approximately 75% of all patients with ARMs have voluntary bowel movements. Approximately 50% have soiling episodes. Such episodes are usually related to constipation; when constipation is treated properly, the soiling usually improves. Approximately 40% of all patients have voluntary bowel movements and no soiling. This is the group which would be described as continent. Definitions of continence based on the Rome Criteria (Ref)—one or fewer accidents per week—have increasingly been used to describe continence. This change will make it easier to compare results between various malformations.

At least 30% of patients with ARMs have fecal incontinence and need a bowel management regimen with a daily enema to keep clean (see Bowel Management). Apart from the anorectal anomaly itself, the status of the sacrum, spine, and muscular development greatly affects a patient's chances of having fecal continence. Even with a perfect reconstruction, a patient with complex malformation, a poor sacrum, or an abnormal spinal cord may not achieve continence.

Potential for continence must be evaluated when the child is older than 3-4 years.

Patients with less complex malformations (eg, rectoperineal fistula or rectal atresia) have excellent outcomes. Girls with rectovestibular fistulas have very good outcomes, except for a tendency to develop constipation. [10] Approximately 60% of boys with rectourethral fistulae and normal sacra have good outcomes. More than 70% of patients with a short-common-channel cloacal malformation and a normal sacrum can develop continence. Patients with very complex malformations (eg, rectobladderneck fistula in boys and long-common-channel cloacal malformations in girls) have poor outcomes.

The sacrum is a good predictor of outcome, in that it correlates with the overall development of the pelvis, including the sphincter muscles and pelvic nerves. Patients with a normal sacrum are much more likely to have fecal continence; those with a hypodeveloped sacrum are much more likely to be incontinent.

Determination of the she sacral ratio allows more objective assessment of the sacrum (see the image below). Patients with a sacral ratio of less than 0.4 rarely achieve continence. A hypodeveloped sacrum is also a good predictor of associated spinal and urologic abnormalities. Sacral x-rays should not be taken in the first 3 months of life, because the coccyx is not yet calcified, and obtaining sacral radiographs during this early period may lead to underreading of the sacral ratio.

Calculation of sacral ratio. Calculation of sacral ratio.

A method that uses the sacral ratio and other variables to assess the likelihood that a patient with an ARM will have continence is outlined in Table 2 below.

Table 2. Numerical Scoring of Chances of Having Fecal Continence for Patient With Anorectal Malformation (Open Table in a new window)

Type of ARM Score
Perineal fistula 1
Anal stenosis 1
Rectal atresia 1
Rectovestibular fistula 1
Rectobulbar fistula 1
ARM without fistula 1
Cloaca < 3 cm common channel 2
Rectoprostatic fistula 2
Rectovaginal fistula 2
Rectobladderneck fistula 3
Cloaca ≥3 cm common channel 3
Cloacal exstrophy 3
Spine Score
Normal termination of conus (L1-2) 1
Normal filum appearance 1
Abnormally low termination of conus (below L3) 2
Abnormal fatty thickening of filum 2
Myelomeningocele 3
Sacrum Score
Sacral ratio ≥0.7 1
Sacral ratio ≥0.4, ≤ 0.69 2
Hemisacrum 2
Sacral hemivertebrae 2
Presacral mass 2
Sacral ratio < 4 3

ARM = anorectal malformation.

Individual scores for ARM type, spine, and sacrum are added together. Total scores are interpreted as follows:

  • 3 or 4 - Good potential for continence
  • 5 or 6 - Fair potential for continence
  • 7, 8, or 9 - Poor potential for continence

A child's outcome may be predicted reasonably accurately. Both patients with good prognoses and those with poor prognoses can be identified. Patients falling between these two categories, however, have proved more difficult to identify, and this issue is being examined in a multicenter study currently under way. Parents can be realistically informed of their child's potential for bowel control, within the first year of life. This avoids a great deal of frustration later in life. Establishing the functional prognosis early is vital to avoid raising false expectations in the parents.

Once the diagnosis of the specific defect is established, the functional prognosis may be predicted. The status of the spine, sacrum, and perineal musculature will affect the counseling given to the parents.

If a given defect is one that carries a good prognosis (eg, rectovestibular fistula, rectoperineal fistula, rectal atresia, rectourethral bulbar fistula, or ARM without fistula), the child may be expected to potty train by age 4-5 years. Such children require supervision and treatment to avoid fecal impaction, constipation, and soiling.

Certain defects indicate a poor prognosis, such as a complex cloaca (common channel >3 cm with a short urethra) or a rectobladderneck fistula. Parents should be informed that the child may require a bowel management program to remain clean. The program should be implemented at age 3-5 years (see Bowel Management).

Patients with rectoprostatic fistulas have an almost equal chance of continence or incontinence. Toilet training should be attempted at age 3-5 years; if this is unsuccessful, a bowel management program can be initiated, depending individual home circumstances. Each year, during vacation, potty training should be attempted; if this is unsuccessful, bowel management should be restarted. As the child grows older and more mature, the likelihood of achieving bowel control improves. Once it is determined that a daily enema is needed, those patients can continue rectal enemas, if these are well tolerated, or enemas can be given via a Peristeen device or in an antegrade manner via a Malone appendicostomy. [11]

Urinary incontinence occurs in boys with ARMs only when they have an extremely defective or absent sacrum or an abnormal spine or when the basic principles of surgical repair are not followed and important nerves are damaged during the operation. The vast majority of boys have urinary control. This is also true for girls, with the exception of the group with complex cloaca, who not infrequently need clean intermittent catheterization (CIC; see Cloacal Malformations).

Careful, regular observation is necessary in these patients for accurately reassessment of their prognosis and for avoidance of problems that can dramatically affect their ultimate functional results. In addition, this group also requires close follow-up of their kidneys to help mitigate the development of renal impairment by actively treating common problems such as recurrent UTIs and reflux and by ensuring good bladder emptying.