Use of MRI in evaluation of anorectal anomalies

Use of MRI in Evaluation of Anorectal Anomalies By Todd M. Sachs, Harry Applebaum, Touraj Touran, Paddy Taber, Ani Darakjian, and Patrick Colleti Los Angeles, California 9 Congenital anorectal anomalies present with a wide spectrum of anatomical deformity. The level of atresia, the extent of the development of the pelvic musculature, and the presence of associated anomalies all greatly influence treatment options. Further complexity may be introduced by previous attempts at correction. Magnetic resonance imaging (MRI) recently has been suggested as an effective tool in evaluation of such patients for surgical repair. Thirteen patients undergoing evaluation for either primary or secondary operations for imperforate anus w e r e studied with MRI. Age range was newborn to 26 years. These patients had anorectal deformities in various stages of t r e a t m e n t - - f r o m newly diagnosed imperforate anus in the newborn to patients who had undergone previous surgical repair of their deformity with poor results. In newborns, MRI was very useful in assessing the level of atresia and determining whether perineal repair was advisable. In older patients with high atresias, MRI was useful in planning operative strategy and predicting the outcome by providing information about the pelvic musculature. This information was directly related to the size of the patient, me, the larger, and therefore the older the patient, the more detail MRI provided about the pelvic musculature. Especially in secondary cases, this information proved to be helpful in determining the cause of the initial failure, me, lack of adequate pelvic musculature or poor placement of the distal colon in relation to the striated muscle complex at the initial operation. MRI demonstrated previous unsuspected lesions such as tethered cord, sacral deformity or urinary tract abnormalities. In newborns with imperforate anus, the meconium inside the colon was hyperintense under MRI, clearly demonstrating the extent of the malformation, obviating the need for contrast agents. MRI is a ne w modality that in a single setting can provide multiplanar imaging without ionizing radiation. It is rapidly emerging as a powerful diagnostic tool in evaluating patients with anorectal anomalies. 9 1990 by W.B. Saunders Company,

patients. The patients were grouped into three categories based on age and any previous surgical procedures (Table 1). All patients received magnetic resonance imaging (MRI) of the abdomen and pelvis preoperatively. Five-millimeter or 10-mm images were generated in the coronal, sagittal, and transverse planes. Anesthetic agents varied according to age. Chloral hydrate (90 to 100 mg/kg) was used for newborns and infants. In older patients thiopental per rectum (30 mg/kg) was used. All patients were monitored while in the scanner for respiratory rate and oxygenation with a pulse oximeter. In newborns, factors analyzed with MRI included the relationship of the level of atresia to the SMC, and the presence of fistulae and associated anomalies. If the distal rectal pouch was at or above the upper border of the muscle complex on sagittal images it was considered to be a high anomaly. If it had passed through the complex, a low anomaly. 3 High and intermediate anomalies were treated with initial colostomy and low lesions with a perineal procedure. Associated anomalies detected included those of the esophagus, heart, genitourinary tract, spinal cord, vertebral column, and bony pelvis. In older infants who had undergone previous colostomy but not MRI studies, the SMC was investigated for size, symmetry, and relationship to the distal bowel. These patients were also studied for the presence of fistulae and other unidentified associated anomalies, particularly those of the spinal cord. In patients presenting with incontinence following a prior attempt at definitive repair, the bulk and symmetry of the SMC was evaluated, as was the positioning of the bowel within the sphincteric mechanism.

RESULTS

A

Three of the 13 patients studied were newborns. In each patient the anatomical relationship of the most distal portion of the bowel to the S M C was adequately visualized with transaxial images. This visualization was aided by the impacted meconium found in the distal pouch, which served as an excellent MRI contrast agent. This property of the meconium can be attributed to the high lipid content of the meconium that enhances the MRI signal. Although the muscle complex was easily identified, the small size of the patients did not allow the fine resolution of individual

From March t987 to March 1989, 13 patients with anorectal anomalies were evaluated. There were 10 male and three female

From the Departments of Pediatric Surgery and Pediatric Radiology, Kaiser Permanente Medical Center and LAC/USC Medical Center, Los Angetes, CA. Presented at the 22nd Annual Meeting of the Pacific Association of Pediatric Surgeons, Portland and Sun River, Oregon, May 22-26, 1989. Address reprint requests to Harry Applebaum, MD, Kaiser Permanente Medical Center, 1526 N Edgemont St, Los Angeles, CA 90027. 9 1990 by W.B. Saunders Company. 0022-3468/90/2507-0029503.00/0

Journal of Pediatric Surgery, Vol 25, No 7 (July), 1990: pp 817-821

817

INDEX WORDS: Magnetic resonance imaging (MRI); anorectal anomalies.

C C U R A T E evaluation of anorectal anomalies is frequently difficult. There are no physical findings or diagnostic modalities that give consistently correct assessment of the level of neonatal atresia, development of the striated muscle complex (SMC), and the presence of associa.ed anomalies. With the advent of the sacroperineal approach to repair, the surgeon has the potential of tailoring the muscles of continence to the requirements of individual anatomical variants. 1'2 MATERIALS AND METHODS

818

SACHS ET AL Table 1. Three Categories of Patients Based on Age and Previous Surgery Group Newborns

Age

MRI Findings

--

Rectal pouch above SMC, no anomalies Cloaca Rectal pouch below SMC, anterior fistula SMC normal size and symmetry, rectourethral fistula SMC normal size and symmetry, anterior fistula SMC normal symmetry, tethered cord, and lipoma of sacral cul-de-sac Levator ani with normal symmetry, a hypoplastic external sphincter Tethered spinal cord, bladder polyps, sacral dysplasia, and hypoplastic sphincter complex Rectum placed anterior and to the right of sphincter complex, mesenteric fat within the muscle complex Severe sacral dysgenesis, hypoplastic SMC with good placement of bowel within muscular sling Hypoplastic pelvis and SMC Rectum placed anterior and to right of sphincter complex, rectourethral fistula Rectum placed anterior and to right of sphincter complex, mesenteric fat within the muscle complex

--Infants for primary pullthrough

9 mo g mo 15 mo

16 mo

Secondary operations after previous pull-through

3 yr

3 yr

4 yr

5 yr 17 yr

26 yr

Outcome Diverting colostomy

Diverting colostomy Anorectoplasty, excision of fistula deVries-Pena with good continence deVries-Pena with good continence deVries-Pena with good continence deVries-Pena with good continence deVries-Pena with improvement in continence

deVries-Pena with good conti-

nence

End colostomy with Hartmann

pouch

No procedure performed deVries-Pena with good continence deVries-Pena with questionable improvement in continence

Abbreviations: MRI, magnetic resonance imaging; SMC, striated muscle complex.

muscles noted in older patients. This did not affect determination of the level of atresia. One infant classified as a low atresia was found to have a fistula traversing the anterior portion of the sphincteric mechanism (Fig 1A). He underwent a perineal procedure with placement of the rectum fully within the sphincteric mechanism, with excellent results. In the other two newborns, MRI demonstrated the rectal pouch to be above the muscle complex and both patients underwent diverting colostomy (Fig 1B). In the four primary pull-through patients with high imperforate anus, MRI outlined pelvic musculature in detail. Each patient had a well-developed SMC. One patient had a ureteral obstruction repaired prior to pull-through. Fistulae were visualized in two patients (Fig 2). Posterior sagittal anorectoplasty was performed with good postoperative sphincter function in all cases. The final group of patients included those considered for reoperation following a previously unsuccessful

procedure elsewhere. Because of the relatively large size of these patients, MR[ was particularly successful in demonstrating anatomy in great detail. In three instances MRI showed the rectum to be placed anterior and somewhat to the side of the vertical fibers of the sphincteric complex (Fig 3). There were large amounts of mesenteric fat within the muscle complex. Each patient underwent a sacroperineal exploration. The mesenteric fat was removed and the bowel replaced symmetrically within the muscle complex. The muscle complex was divided tangentially in order to correct the inaccurate placement of the bowel at the previous procedure. Two patients had greatly improved continence. The other patient had only questionable improvement in continence, at least partially due to lack of motivation secondary to Down's syndrome. The fourth patient was a 3-year-old boy with prior perineal anoplasty and incontinence. He had severe sacral dysplasia, a tethered spinal cord, and other genitourinary anomalies (Fig 4). MRI demonstrated a

MRI EVALUATION OF ANORECTAL ANOMALIES

819

Fig 1. (A) Newborn boy with low imperforate anus. M, meconium in bowel; arrow, scrotal perineal fistula. (B) Newborn boy with high imperforate anus. B, dilated obstructed bowel above SMC.

severely hypoplastic sphincter complex on the side of sacral dysplasia. Because of the parent's strong desire to achieve any improvement, a tangential revision anorectoplasty was attempted following repair of the tethered spinal cord. As predicted, there was minimal postoperative improvement. A fifth patient was turned down because of severe sacral dysgenesis, hypoplastic musculature, and multiple failed previous procedures with postoperative complications. MRI showed good placement of the bowel within the hypoplastic sling. This patient underwent a Hartmann procedure. Another patient was advised against further surgery due to a severely hypoplastic pelvis and musculature observed by MRI. DISCUSSION

The ultimate determinant of continence following repair of imperforate anus is the optimal placement of the bowel symmetrically within the SMC. The sacro-

perineal approach of Pena and deVries allows placement with great precision, but the procedure must be timed and modified to fit the anatomical variants encountered. ~,2 Physical examination in the newborn with imperforate anus has often been of questionable reliability. The presence of a perineal fistula indicates a low lesion but may not accurately display the relation of bowel to the anterior portion of the external sphincter. The degree of development of the gluteal crease and the presence of bony abnormalities only suggest the degree of bilateral muscular development. Previously available imaging studies have only been able to hint at the position of the bowel in relation to the SMC by inference (Wangensteen-Rice invertograms) or in a relatively low-resolution pictorial manner (ultrasound and computed tomography [CT] scan). 47 Ultrasound has only been useful for demonstrating renal anomalies. 5 CT scan provides information on musculature, bony

820

SACHS E T A L

Fig 4. Three-year-old boy with a tethered spinal cord ending in a large sascral cul-de-sac liporna (L).

Fig 2. Nine-month-old boy with high imperforate anus being considered for a pull t h r o u g h procedure. C, distal colon; F, fistula between colon and prostatic urethra.

and spinal anomalies, but only in one anatomic plane. The gonads receive a considerable amount of ionizing radiation with CT. 4'7'8 MRI provides safe multiplanar imaging with excellent contrast and resolution. This combination of images provides three-dimensional information easily translated into expected findings at surgery.3'7 In newborns, the level of atresia was accurately depicted in all cases. 3 It was found that meconium serves as an

Fig 3. Twenty-six-year-old man with fecal incontinence. Eccentric position of bowel to the right of sphincter complex (s), and large amount of mesenteric fat within muscle complex (f).

excellent enhancer of the MRI signal, acting as a contrast agent within the bowel. This helped indicate whether or not primary repair was advisable. MRI also helped in patients with low lesions because the knowledge of a rectal fistula passing through the anterior portion of the sphincter indicated that more than a simple cutback procedure was the optimal surgical treatment. With primary pull-through patients, MRI provided information regarding the development of the pelvic musculature that is essential for the surgeon to better plan operative intervention and assess the likely functional outcome. The muscle complex was demonstrated in great detail in older and therefore larger patients, allowing the surgeon to plan the angle to be taken when dividing the SMC during the pull-through procedure. In previously operated on patients with fecal incontinence, MRI provided invaluable information regarding the cause of failure of the initial procedure. 3 Findings included lack of adequate SMC, improper placement of the bowel in relation to this musculature, or large amounts of mesenteric fat that had been brought down within the SMC, thereby compromising the sphincteric function.3 This information greatly decreased the amount of dissection needed to adequately assess abnormal anatomy and thus minimize muscular damage during the corrective procedure. In some cases, in which severely hypoplastic musculature

MRI EVALUATION OF ANORECTAL ANOMALIES

821

was d e m o n s t r a t e d unilaterally, or in those in which it was evident t h a t t h e r e was no technical p r o b l e m but r a t h e r a deficiency of m u s c u l a t u r e , no further s u r g e r y was advised. T h e M R I e x a m i n a t i o n surveys these patients for a n u m b e r of associated a n o m a l i e s in both newborns a n d older infants. These included those of the digestive and g e n i t o u r i n a r y t r a c t s a n d t e t h e r e d spinal cords. It is i m p o r t a n t to correct the spinal cord a b n o r m a l i t i e s

early in life, and c e r t a i n l y prior to p u l l - t h r o u g h procedures, to m a x i m i z e innervation and, therefore, size and function of the sphincteric m u s c u l a t u r e supplied by the sacral nerve roots. M R I provided previously u n o b t a i n a b l e benefits in all categories of patients studied. W h e n c o m b i n e d with a thorough physical e x a m i n a t i o n , it will become preferred and usually the only i m a g i n g s t u d y necessary in the preoperative evaluation o f a n o r e c t a l anomalies.

REFERENCES

1. deVries PA, Pena A: Posterior sagittal anorectoplasty. J Pediatr Surg 17:638-643, 1982 2. Pena A, deVries PA: Posterior sagittal anorectoplasty: Important technical considerations and new applications. J Pediatr Surg 17:796-811, 1982 3. Pringle KC, Sato Y, Soper RT: Magnetic resonance imaging as an adjunct to planning an anorectal pull-through. J Pediatr Surg 22:571-574, 1987 4. Kohda E, Fujmioka M, [kawa H, et al: Congenital anorectal anolmaly: CT evaluation. Radiology 157:349-352, 1985

5. Oppenheimer DA, Carroll BA, Schochat S J: Sonography of imperforate anus. Radiology 148:127-128, 1983 6. Kuvlander G J: Roentgenology of imperforate anus. AJR 100:190-201, 1967 7. Mezzacappa PM, Price AP, Haller JO, et al: MR and CT demonstration of levator sling in congenital anorectal anomalies. J Comput Assist Tomogr 11:273-275, 1987 8. Smith ED: The bath water needs changing, but don't throw out the baby: An overview of anorectal anomalies. J Pediatr Surg 22:335-348, 1987