Significance of the ‘line sign’ in the diagnosis of congenital imperforate anus on prenatal ultrasound

Imperforate anus is a common congenital malformation that is caused by hindgut development defects or retardation. Imperforate anus occurs in approximately 1 per 5000 live births [1]. The deformity can occur alone or with other congenital malformations or chromosomal abnormalities. The prenatal diagnosis of imperforate anus is difficult, as there are no direct ultrasound signs. The lack of the ‘target sign’ is often used for predictive diagnosis. In previous studies, the diagnosis of imperforate anus by prenatal ultrasound was based on indirect sonographic features, such as dilatation of the proximal bowel with atresia and the enterolith sign [1‐3]. However the prenatal detection rate of anal atresia is yet low (8.2%) [4].

Prenatal diagnosis of imperforate anus is not always possible. Nevertheless, consciousness of the condition and the ability to recognize the most typical ultrasound findings in imperforate anus may improve the detection rate. This study compared sonographic features of the anus between normal foetuses and foetuses with imperforate anus. The aim of this study was to describe the imaging features of imperforate anus by prenatal ultrasound. The secondary aim was to explore whether the ‘line sign’ can be used as a secondary assessment to determine the type of the malformation following non visualization of the ‘target sign’.

During the 3-year study period, through the ultrasound examination of 16,475 pregnant women, 19 fetuses presenting with imperforate anus were prospectively evaluated for the absence of the ‘target sign’ and the presence of the ‘line sign’. Table 1 shows the relevant sonographic features in these cases. There were 13 (68.4%) males and 6 (31.6%) females among the 19 cases of imperforate anus. There were 18 (94.7%) cases of non-dilatation and 1 (5.3%) case of dilatation. There were 3 (15.8%) cases of polyhydramnios, 1 (5.3%) cases of oligohydramnios and 15 (78.9%) cases of a normal amniotic fluid index.

There were 3 (15.8%) cases of simple imperforate anus and 16 (84.2%) cases of imperforate anus with other malformations. The other malformations included 5 cases of VACTERL syndrome, 3 cases of OEIS complex, 2 cases of urorectal septal sequence syndrome, 3 cases of caudal regression syndrome, 1 case of malformation of the right auricle, 1 case of gallbladder agenesis and 1 case of horseshoe kidney with a single umbilical artery. Sonographic findings were compared with clinical results, as shown in Table 2.

Among the 16,475 fetuses in which the regular the ‘line sign’ was identified prenatally, no cases of imperforate anus were reported postnatally or at the time of postmortem examination. Of the 19 cases of imperforate anus, 16 cases were diagnosed by the ultrasound image feature called the ‘line sign’. 1 case with tail degeneration was low type imperforate anus with the irregular ‘target sign’ not a real ‘target sign’. There was two false-negative case, in which the ‘target sign’ was found, but irregular. Both cases were found to be isolated low imperforate anus after delivery. Overall, in our study, absent the ‘target sign’ and present of the ‘line sign’ on prenatal sonography had false negative rate of 10.5% and no false positive cases for the diagnosis of imperforate anus.

In 16 cases, we were able to confidently demonstrate the imperforate anus by identifying the absence of the ‘target sign’ and the presence of the ‘line sign’. There was one case of low type imperforate anus with caudal regression syndrome. Postmortem examination of the fetus showed no anal canal. The distance between the blind end of rectum and anal skin was 0.5 cm. Pathology showed the existence of anal column, anal mucosa, irregular internal sphincter and external sphincter. The development of anal canal in low type imperforate anus was closer to normal.

The results of serological screening were normal in 19 cases. Non-invasive prenatal testing (NIPT) was performed in 6 cases. The result was normal in 5 cases and abnormal in 1 case. In the abnormal case, 18q22.1 - q23.79 m deletion and 11.79 mb microduplication were detected.

The prenatal diagnosis of imperforate anus is difficult. However, the diagnosis may be suspected by the sonographic observation of colon dilatation in or beyond the second trimester or by the absence of a typical anal sonographic appearance in the third trimester. Often, the diagnosis is only made after birth [6, 7]. Most previous studies have been case reports [8].

Lianli et al. reported that imperforate anus can be diagnosed according to the area of the anal sphincter, the distance between the anal sphincter and ischia, absence of the anal canal and the anterior and posterior size of the rectum [9]. Brantberg et al. reported the prenatal sonographic diagnosis of imperforate anus relied on indirect findings, such as abnormally dilated distal bowel segments in early second trimester or the presence of calcified intraluminal meconiumor enterolithiasis in the second and third trimesters [1]. However, our study found that bowel dilatation is not present in most cases of imperforate anus, and there are no signs of enterolith or hydramnios. Therefore, it is difficult to diagnose imperforate anus with the above methods.

Ying et al. studied the cross-sectional features of the anal sphincter in normal foetuses on ultrasound examination (‘target sign’) [5]. Few studies have summarized the sonographic features of the anal sphincter in foetuses with imperforate anus. Ochoa et al. have found that in a high-risk population, the absence of perianal muscular complex (PAMC) seems to be a highly sensitive and specific sonographic marker for anorectal atresia. But the role of routine sonographic identification of the PAMC for cases of isolated imperforate anus remains to be determined [10]. In this study, by analysing a large number of ultrasound image features of the foetal anus, we found a typical ultrasound image feature of imperforate anus (‘line sign’).

In this study, we found that sonographic observation of the imperforate anus revealed the ‘line sign’ (Fig. 2a), with no hyperechoic area or round hypoechoic ring. The reflection of the interface between the outer layer of the anal sphincter and surrounding tissues was absent, and the area of the anus was completely covered with skin. When the direction of the ultrasound beam was perpendicular to the gluteal sulcus, a hyperechoic line formed in the area of the anus. We denoted this sonographic feature as the ‘line sign’. Therefore, we regarded the ‘line sign’ formed when the direction of the ultrasound beam was perpendicular to the gluteal sulcus as the diagnostic standard for imperforate anus. After autopsy, we also found that the higher the position of the imperforate anus was, the more obvious the ‘line sign’ (Fig. 2). There was no anal canal. The anal sphincter was absent or maldeveloped.

Rohrer et al. pointed out that the absence of the ‘target sign’ is the foremost direct imaging sign suggesting anorectal malformation [11]. However, we found that there was a irregular ‘target sign’ in low imperforate anus. Imperforate anus can be divided into high type and low type depending on the relation-ship between the distal rectal pouch and the puborectalis muscle [12]. Haber et al. have shown that the prenatal diagnosis of low type imperforate anus on ultrasound is very difficult [13, 14]. There were 3 cases of low type imperforate anus in our study. Prenatal examination showed no bowel dilatation. There were a short ‘line sign’ easily overlooked and a irregular ‘high-low-high’ concentric circular echo, which looked like the ‘target sign’ (Fig. 3). There were two layers of smooth muscle in the muscular layer of the rectal wall, one for the inner ring and one for the outer longitudinal layer. This was consistent with the arrangement of the sphincter of the anal canal. Therefore, there was also a hypoechoic ring on sonographic observation. The rectal wall was thinner than the sphincter of the anal canal. Therefore, the hypoechoic ring formed by the rectal wall was thinner than that of the ‘target sign’. After the induction of labour, a sagittal scan with a high-frequency probe clearly showed the intestinal wall structure of the blind end of the rectum (Fig. 4). In general, the ‘high-low-high’ concentric circle echo formed by low type imperforate anus was different from that formed in normal foetuses (‘target sign’).

Following the analysis of the medical records and results from our series, we developed a clinical-diagnostic flowchart to be used in cases of prenatally detected congenital imperforate anus (Fig. 5) .

The ability to recognize the most typical ultrasound findings in imperforate anus can improve the detection rate. The absence of the ‘target sign’ and then the presence of the ‘line sign’ can assist in the diagnosis of imperforate anus. The ‘line sign’ can be used as a secondary assessment to determine the type of the malformation following non visualization of the ‘target sign’.

The higher the position of the imperforate anus is, the more obvious the ‘line sign’. It is worth noting that the finding of the short ‘line sign’ and irregular ‘target sign’ can not ignore the low type imperforate anus. Several additional limitations apply to our study. These include that the small sample size of low type imperforate anus, and we didn’t explore the role of 2D vs. 3D ultrasound with regard to the prenatal diagnosis of the imperforate anus.